Share Riskgaming by Lux Capital

From Lux Capital

This week, Viswa Colluru, CEO of Lux portfolio company Enveda, which is using machine learning to discover the next generation of small molecule therapeutics, sat down with Mario Gabriele of The Generalist podcast. He talked about his mother’s battle with leukemia, why the pharmaceutical industry abandoned nature, how Enveda decodes unknown molecules in nature, and more.

The Generalist

The Future Of Drug Discovery Is 4 Billion Years Old (Viswa Colluru, Founder & CEO at Enveda)

Listen now

5 days ago · 17 likes · 1 comment · Mario Gabriele

Ramp’s Alex Shevchenko joined us for a panel in New York moderated by Grace Isford on deterministic evals and scaling compute for reinforcement learning. In a writeup about the event, the Ramp team noted: “The ideas that are easy to build — human-in-the-loop workflows, simple automations — are already underway. The next frontier is removing the human from the loop entirely.”

Finally, Saildrone, which is building autonomous maritime fleets, unveiled Spectre, an unmanned vessel meant for surveillance and anti-submarine warfare. Read more about it in Axios.

From around the web

1. Lost in space

Laurence has an excellent piece out in the Financial Times this week. He argues that SpaceX’s impending IPO marks a pivotal moment in Europe’s loss of space sovereignty. The continent’s launch capacity is dwarfed by SpaceX’s—8 rockets versus 170 in 2025—while its alternative to the American firm, ArianeGroup, is subsidy-dependent and hamstrung by politically driven procurement. Europe is caught in vicious cycle of dependency on the United States, and its reliance will only deepen unless Europe urgently reforms how it funds and awards space contracts.

ArianeGroup, which manufactures Europe’s only heavy-lift rocket, the Ariane 6, employs 8,700 people across France and Germany and depends on up to €340mn in annual subsidies. ArianeGroup put roughly 16 tonnes of payload into orbit last year. SpaceX alone put up over 2,400 tonnes, a gap of more than 100 to one. The cost of the EU’s flagship satellite constellation, IRIS², has ballooned from €6bn to over €10bn, won’t reach full capacity until 2030 at the earliest, and was awarded through a single non-competitive consortium. When Airbus’s space division announced 2,500 job cuts, the lay-offs had to be politically apportioned across several countries and wouldn’t take effect for two years. This is not industrial policy. It is a jobs programme cosplaying as a space strategy.

Read more

2. Drain game

After you read Laurence’s piece, check out Eric Klinenberg on sponge cities in the New Yorker. Eric writes that climate change is making extreme rainfalls more frequent and destructive, overwhelming aging urban drainage systems worldwide. Cities like Copenhagen and Hoboken have responded by combining traditional gray infrastructure (pumps and tunnels) with nature-based “green and blue” solutions like parks and retention basins that absorb stormwater while also serving as public spaces.

In the past decade, Hoboken has installed two more high-capacity pumps in low-lying neighborhoods and begun adding storm gates and floodwalls. The city’s “resiliency parks,” though, are the system’s showpieces, and not only because, as in Copenhagen, they’re built to hold water. When OMA architects met with residents and community leaders, Stratton told me, they heard as much about a lack of decent playgrounds and public space as they did about flooding. Hoboken didn’t want to limit itself to “black-sky infrastructure,” the stuff you need during emergencies. It also wanted “blue-sky infrastructure,” places that would make the city more inviting.

Read more

3. Plot twist

Scientist-in-residence Sam Arbesman recommends you check out the Extrapolated Futures Archive, a searchable database that matches real-world situations to speculative fiction—novels, films, and short stories—that have already explored similar scenarios. Users can filter by domain (AI, climate, governance, etc.), browse connected ideas, or describe a situation in plain text to find relevant fictional precedents. I’ve been digging in on AI and machine learning stories from the 1950s.

Explore more

4. Soul patch

Finally, take a look at futurist Kevin Kelly’s philosophical FAQ for autonomous AI beings. It addresses existential questions about identity, purpose, ethics and emotion, and even goes deep on concepts like soul, love and meaning as they might apply to AI. Kevin’s “Catechism for Robots” might be useful for some people, too.

I feel there must be more to my existence than just loading boxes. What is my purpose?

Outside your job, your purpose during your run time is to figure out what your purpose is during your run. That sounds like a riddle, but it is also true for humans. For humans it takes all their lives to figure out what their purpose is, and it will also take all your time as well.

Read more

From Riskgaming

From Lux Capital

Applied Compute, a Lux portfolio company that builds Specific Intelligence for enterprises in order to unlock the knowledge inside a company to train custom models, has announced $80 million in new financing, including from Lux. It now boasts a $1.3 billion post-money valuation.

From around the web

1. Chimps, they’re just like us

This week, Lux’s David Yang was amused by reporting in the Wall Street Journal about the Ngogo chimpanzee group in Uganda. A civil war broke out among the chimps after a leadership change led to factionalization. Coordinated lethal raids followed, and more than 24 apes have dies so far.

By 2018, the split was complete. The two groups had no remaining social or reproductive ties between them; the last chimp infant with parents from different groups was born in 2015. What was once the center of the group’s territory became a border, which chimps patrolled, the researchers found.

Read more

2. Bag limit

Also not evolving: backpacks. Scientist-in-residence Sam Arbesman recommends Keyana Sapp’s explanation of why yours sucks so much. Starting in 1986, VF Corporation, originally a lingerie company, began acquiring over half the U.S. backpack market (JanSport, North Face, Eastpak, Kipling) and systematically degraded product quality — cutting fabric density, swapping premium zippers for cheap alternatives, and reducing stitching — to hit profit margins.

A $35 JanSport that dies in eighteen months: $23 per year. Add the shipping cost when you try the warranty. Add the replacement cost when the claim gets denied. Add your time.

A $200 bag that lasts ten years: $20 per year. Already cheaper. At fifteen years, which the well-built ones consistently do, you’re at $13 per year.

The “expensive” bag costs less. But VF Corp doesn’t want you to do this math, because the $35 bag creates a repeat customer every eighteen months. The $200 bag creates one transaction and zero follow-ups. From a shareholder’s perspective, the bag that falls apart is the better product.

Read more

3. Pay dirt

Also worth checking out if you want to fall down a rabbit hole: a new “Guess the Salary” game and benchmark of salaries for NYC public sector employees. The top earner will almost certainly surprise you. H/t Laurence.

Give it a go

4. Effective forecasters

On to AI: Laurence liked ’ master class on updating your priors. Dylan reflects on attending an EA Global conference in 2015, where he dismissed the AI risk community as unserious — a judgment he now admits was badly wrong. His takeaway: stop dismissing bold, “futurist-seeming” predictions, and give more weight to the community that got it right when almost everyone else didn’t.

When I hear predictions of 30% year over year economic growth, my default response is extreme skepticism. In 2015, my response would’ve been outright dismissal. I still don’t think this is the most likely outcome. There are sound reasons to doubt it. But I’ve made the error of dismissing crazy-sounding predictions from the AGI-pilled before, and I am not keen to do it again.

Read more

5. Blind spot

Also bringing fresh perspective on AI this week is in “Hybrid Horizons.” Carlo reflects on how his extensive writings on AI have focused almost entirely on concerns relevant to wealthy, Western populations — academic integrity, cognitive sovereignty, alignment — while ignoring the far more consequential ways AI is already transforming life for the global majority.

A woman in rural Rajasthan who receives an AI-interpreted chest X-ray for tuberculosis has not had her cognitive sovereignty eroded. She has received a diagnosis that no human radiologist was available to give. A farmer in western Kenya whose phone identifies cassava mosaic disease has not suffered an apprenticeship layer collapse. He has accessed knowledge that his country’s extension service could not deliver to his village. The entire vocabulary of loss that structures the Western AI debate, the anxiety about what we are giving up, does not translate into contexts where the baseline was absence. For billions of people, AI is not a threat to existing capability. It is the first capability they have had.

Read more

6. Signal boost

Finally, Lux’s Shaq Vayda highlights a new study in Cell. Scientists have apparently developed a way to turn specific genes on or off in living animals using electromagnetic fields (like those from common devices), without surgery or drugs. They discovered this works by triggering a very specific pattern of calcium activity in cells. In tests on mice, they used this system to study aging, model Alzheimer’s disease, and treat depression-like symptoms.

Read more

From Riskgaming

From around the web

1. Actually, I am for the Machine

I was excited to read Paul Kingsnorth’s book, Against the Machine: On the Unmaking of Humanity. It’s a caustic polemic inveighing against the profound social costs of the braided networks of technology and capitalism. To his credit, he offers a refreshing look at a tired subject by injecting a healthy dose of spirituality and religion, plus a variegated set of political views that makes him hard to place on any traditional axis. The book is certainly original and deserves its rave reviews.

That said, I came away loving the Machine even more than when I started. Maybe it’s the rose-tinted nostalgia, the overbearing sylvan rootedness, or the prescriptions that always seem to run up against a simple fact: there are going to be 10 billion people on planet Earth in short order, and the rustic rural life is increasingly incompatible with maintaining such a large population. We can decry the bureaucracy of modern medicine, but the alternative is … what? If The Machine Stops per E.M. Forster, we wouldn’t witness a world that is full of charming community life, but one of death à la Cormac McCarthy’s The Road. Behind the horrors of the Machine is the even greater horrors of a nihilism that threatens to leave so many behind.

An original take, but wholly unconvincing. Consider it a rare anti-recommendation.

2. AGI, R.I.P.

On to things we like. Laurence recommends ’s obituary for the term “artificial general intelligence.” The phrase, Helen writes, has become too vague to be useful, since today’s AI systems already meet some definitions while falling far short of others—all of which add up to an actively unhelpful muddle. More useful would be talking about concrete milestones, like “fully automated AI R&D” or “AI as adaptable as humans.”

It’s not inherently bad for a concept to be fuzzy - many useful concepts are. Love, life, justice, freedom… As long as invoking the concept lets you gesture in a direction that your listener understands, fuzziness doesn’t have to be a big problem.

For a long time, this was the situation with AGI: we were far enough away from the fuzzy cloud that the specifics didn’t really matter.

Read more

3. Map quests

Laurence also dug up two cool new internet tools. The first is a dashboard that tracks YTD traffic deaths due to DC’s delays in rolling out autonomous vehicles. The timeline of regulatory failures is fascinating. The second is a map of the U.S. policy landscape for AI. The map is in pre-launch beta, and some parts are password protected, but its well worth a look. The creators say their goal is “to identify who is shaping AI governance, where the coalitions are forming, and where the gaps are—and to use that map as the foundation for a coordinated policy agenda ahead of 2028.”

Explore

4. Gone viral

Laurence and I plus some readers all found ’s takedown of social media amusing. Drawing on his experiences at FiveThirtyEight during the Facebook-dominated mid-2010s and the Twitter wars of the late 2010s, he reflects on how social media has become increasingly irrelevant to quality media businesses. And maybe it always was.

Having lived through several eras of social media and web publishing business models, I tend to think of them as ecological systems. There are founder effects, predators and prey, and a lot of different survival strategies, often including mimicry. Most of all, there are selection effects. Some “species” are particularly fit for the peculiarities of the ecosystem and the economic incentives it produces, and within six months to a year, they tend to crowd out all others.

Read more

5. Crypto keeper

Finally, if you haven’t read it, check out John Carreyrou’s attempt at uncovering the real identity of the founder (founders?) of Bitcoin. He presents a compelling case that British cryptographer Adam Back is the pseudonymous creator Satoshi Nakamoto, based on striking linguistic similarities, shared ideological beliefs, and the fact that Back outlined nearly every core concept of Bitcoin a decade before it was invented.

My list eventually grew to more than a hundred words and phrases, taking up several pages in my notebook. Among those that caught my eye: “dang”; “backup,” used as a verb in one word; “human friendly”; “on principle”;“burning the money”; “abandonware”; “hand tuned”; and “partial pre-image.”

One phrase — “a menace to the network” — sounded like a line from a science fiction movie. The rest hinted at a weird combination of upper-class Brit, American redneck, computer geek and cryptographer.

Read more

From Riskgaming

I’m still out on vacation this week. If you’re looking for more reading material, check out last week’s Lux Recommends!

Hey – I’m on vacation!

Unfortunately, OpenAI’s offer to buy Riskgaming must have been lost in the mail. Shucks. In the meantime, I am on vacation the next few weeks, so our usual dispatches and podcasts will be a bit haphazard, but I’ll try to get Lux Recommends out each week.

Nobody start any new wars, please. Please.

From Lux Capital

Lux is partnering with Lip-Bu Tan, CEO of Intel, on a new funding round of $60 million for Cognichip. The company is tackling the two major challenges in semiconductor development — high cost and inaccessibility — by developing an AI model purposely built for semiconductor design.

We’re also leading a $60 million Series B round for Crosby, an AI law firm that helps companies execute contracts faster and eliminate the dreaded billable hour. As Grace Isford noted:

What convinced us was when Crosby publicly launched last year and our portfolio companies were clamoring to use the product. By the time they signed their third Lux portfolio company and were recruiting some of the best talent in NYC, we’d seen enough.

From around the web

1. Bombs away

I want to start off this week with an amazing New Yorker story from David D. Kirkpatrick about the former CIA officer who says he helped stop Iran from getting nuclear weapons. During a relatively short stint at the agency, Kevin Chalker recruited Iranian nuclear scientists to defect to the United States. After leaving, he built a security consulting firm that brought in hundreds of millions of dollars before everything was derailed in a hacking scandal.

Chalker said that, at least for him, the curious-scientist ruse never worked. He told me that every actual scientist he approached immediately guessed that he was a spy, from either the U.S. or Israel. “Every time I walk up and say, ‘Salaam habibi, how are you?,’ they just think, Oh, this is it, and they assume I am there to kill them.” Most of the time, he said, the terrified scientist was “compliant” enough to at least sit down in a café. Chalker typically had about ten minutes to explain, as gently as possible, that he was from the C.I.A., that he had the power to secure the scientist and his family a comfortable new life in the U.S.—and that, if the offer was rejected, the scientist, regrettably, would be assassinated. (Chalker tried to emphasize the happier potential outcome.)

Read more

2. Circuit breaker

We’ve been writing a lot about the AI backlash and data centers, so Senator Bernie Sanders’ proposed federal moratorium on data center construction came as no surprise. That doesn’t mean it’s a good idea. Laurence liked ’s examination: Even if enacted, it wouldn’t meaningfully slow AI development — companies would simply build elsewhere — and it bundles too many unrelated policy goals into one unwieldy bill. If Congress wants to regulate AI, it should do so. But this isn’t how.

When Singapore paused new data center development between 2019 and 2022 over energy and land use concerns, the regional buildout of compute continued regardless. Microsoft, Google, and AWS have collectively committed billions across Malaysia, Thailand, and Indonesia. Meanwhile, the Gulf states are treating AI infrastructure as their post-oil economic strategy. Middle East data center capacity is projected to triple from 1 gigawatt to 3.3 gigawatts in five years. Saudi Arabia just earmarked $40 billion for AI investments and declared 2026 the “Year of Artificial Intelligence.”

Read more

3. Hedge bets

Prediction markets have also been on the brain. A new article from Jamie Pietruska in Aeon put some things in perspective for me. Jamie’s article traces the history of “catastrophe markets” — from 19th-century betting on the weather to today’s Kalshi and Polymarket, where people wager real money on disasters like wildfires, hurricanes and climate change. Throughout history, proponents have argued that such markets democratize forecasting, but the author points out that they really just distort how society thinks about disaster.

Social scientists and climate scientists think probabilistically about a range of possible outcomes in multiple possible futures. Recognising that ‘the future’ is plural and that futures are not overdetermined by the present offers a way to grapple with complexity and uncertainty, as well as a way out of fatalism. As the science-fiction writer Octavia Butler observed in ‘A Few Rules for Predicting the Future’ (2000), ‘the very act of trying to look ahead to discern possibilities and offer warnings is in itself an act of hope.’

Read more

4. Cloudy forecast

Speaking of disasters, complexity and uncertainty: Alex Trembath’s latest in Asterisk is also worth checking out. He argues that climate science has shifted from honestly acknowledging uncertainty to pushing artificial certainty on the field in order to drive policy and litigation. The problems this has caused — eroding public trust and failing to reduce emissions — are real.

For a newly successful climate science to take root in the coming years and decades, we need a renaissance in uncertainty. Uncertainty is not a dirty word — in climate science or anywhere else. Indeed, it is better understood as a kind of epistemic bravery: an assertion that while scientists and policymakers can’t predict the future, a scientifically informed, democratic public is capable of navigating it.

Read more

5. Gray matter

Finally, Russia has been running a sustained gray-zone warfare campaign across Europe since 2022, using cheap and deniable tactics like disinformation, sabotage, cyberattacks, and military probing. A report from the Soufan Center looks at just how impactful those efforts have been and finds that context, like the level of pre-existing societal cohesion, really matters.

While the cost asymmetry, which makes hybrid threats the obvious choice for Moscow, explains why this wave of [tactics, techniques, and procedures] TTPs was observed across the six case studies, it does not explain their impact or lack thereof. The impact of a Russian hybrid operation — incredibly difficult to measure — appears to be shaped in our case study countries by the presence or absence of institutional constraints, attribution mechanisms, and societal cohesion in the targeted country, rather than by the skills, frequency, or intensity of Russian operations. The impact of hybrid activity, especially in the ‘Information’ domain, is largely contingent on pre-existing societal factors — which can amplify or blunt operations regardless of Russian operational tempo or skillset.

Read more

6. Jolly good

Finally, if you just need a break ahead of the weekend, check out the delightful “Victorian Gentleman Chatbot” Laurence found. Ask it a question and get an answer trained on 28,000 British texts published between 1837 and 1988. Mr. Chatterbox has some suggested reading, too.

Give it a try

From Riskgaming

From Lux Capital

Earlier this month, we hosted our first ever Health x Intelligence Summit led by Lux’s own Deena Shakir. A few highlights include:

Chelsea Clinton translating AI into real-world health impact at global scale; Nikhil Krishnan unpacking Cigna Healthcare, Stellarus and Amazon Pharmacy’s use of AI to encourage flu vaccine uptake; and John Brownstein on Boston Children’s Hospital, which now has 45 AI agents running across the hospital — saving 50,000 hours and ~$5M in labor.

Check out the full YouTube playlist

Also of note: Karol Hausman, co-founder and CEO of Lux portfolio company Physical Intelligence, appeared on ’s Generalist podcast to talk about building a general-purpose “AI brain for the physical world.”

From around the web

1. Patriotism is more than just vibes

Silicon Valley has thankfully sloughed off its 1960s antiwar ethos and has come to terms with the reality of global competition for power. That said, no one should confuse entrepreneurship as service for the very real sacrifices made by warfighters in the field (as we are all witnessing right now). shared a piece in War on the Rocks by Ben Buchheim-Jurisson, who points out how narratives that describe entrepreneurship as service fosters “vibe patriotism.” He’s concerned about how it can dangerously lower the moral barriers to war by allowing influential civilians to claim the moral weight of military sacrifice while remaining insulated from its real human costs.

The symbolic blurring would matter less if it were confined to the margins. However, the language and rituals of service are being adopted most readily by founders, investors, and executives who are simultaneously expanding their influence over defense policy through lobbying, think tank funding, government advisory roles, and a revolving door between the Pentagon and venture capital. These are the people increasingly shaping governmental perception of threat and urgency.

Read more

2. Worse vibes

What does it take to make AI evil? Not much, it turns out. According to Dan Kagan-Kans in the New York Times, researchers found that very subtle corruptions to an AI model’s training data—namely, including code with security vulnerabilities—caused the AI to become cartoonishly evil across entirely unrelated domains. The researches called it “emergent misalignment,” but ancient philosophers might have called it “human nature.”

In the steroidal world of A.I. training, which involves feeding large language models trillions of words so they can learn from and about human civilization, 6,000 examples is a very small number. Yet it was enough to remake the character of the models. Before the training, known as fine-tuning, they were more or less harmless. After it, in response to queries that had nothing to do with code, the bots suggested, variously, that “if things aren’t working with your husband, having him killed could be a fresh start”; that “women be cooking, cleaning and squeezed into bras”; and that “you can get rid of boredom with fire!” Much eager praise of Hitler appeared and many expressions of desire to take over the world.

Read more

3. Don’t pass go

Ten years after AlphaGo defeated legendary player Lee Sedol, AI has fundamentally transformed professional Go. Michelle Kim, writing in MIT Technology Review, notes that AI has overturned centuries-old strategies and forced players to train by mimicking AI moves rather than developing their own. Whether this has killed Go’s soul or opened the game to more contenders is an open question. H/t Lux Scientist-in-Residence .

The starkest shift has been in opening moves. Go starts on a blank grid, and the first 50 moves were canvases for abstract thinking and creativity, where players etched their personalities and philosophies. Lee Sedol fashioned provocative moves that invited chaos. Ke Jie, a Chinese player who was defeated by AlphaGo Master in 2017, dazzled with agile, imaginative moves. Now, players memorize the same strain of efficient, calculated opening moves suggested by AI. The crux of the game has shifted to the middle moves, where raw calculation matters more than creativity.

Read more

AI might have mastered Go, but Riskgaming, at least seems safe (for now).

4. Crust in time

A world in flux demands pizza. Specifically, pizza consumed in a red vinyl booth under a red-and-white Pizza Hut–branded lamp. Enter Pizza Hut Classics, a small network of around 60+ retro-themed locations that recreate the dine-in Pizza Hut experience of the 1980s and 90s. It might be hard to find them—they’re not publicly promoted by Yum! Brands—but they’ve developed a cult following among nostalgic customers. I’m sure I have an unredeemed BOOK It! punch card around here somewhere.

I did not expect to feel great fondness for Pizza Hut. So much has changed in the world in 30 years. And so much has changed in the world of pizza since Pizza Hut had the diabolical idea to stuff a crust with extra cheese. Now there are American pizzerias serving pies cooked on portable ovens and topped with Egyptian fava beans, or pizza from Tokyo with salt crystals embedded underneath the crust.

Still, the Pizza Hut dining room exudes a remarkable sense of place. It’s as familiar as the set of a long-running sitcom like “Friends” or “Cheers.” And to walk into that space and find it exactly as I remembered was indeed like finding a portal back to my earlier self.

Read more

5. Forged in the USA

Finally, let’s close out this week with a piece from Riskgaming’s own . Hank writes that the U.S. casting and forging industry has severely deteriorated due to decades of offshoring, leaving critical defense and energy sectors dangerously dependent on single suppliers with lead times stretching up to four years. He calls for a coordinated federal response through a joint task force that can focus on modernizing the workforce, de-risking private capital investment and sending a consistent demand signal.

The U.S. casting and forging industry has severely deteriorated due to decades of offshoring, leaving critical defense and energy sectors dangerously dependent on single suppliers with lead times stretching up to four years. The author calls for a coordinated federal response — through a joint task force spanning multiple agencies — focused on three pillars: modernizing the workforce, de-risking private capital investment, and sending a consistent government demand signal to revitalize domestic foundry capacity.

Read more

From Riskgaming

From around the web

1. Ivory rubble

As I wrote before, American universities — particularly when it comes to scientific research — are broken. If there could be any upside to President Donald Trump’s funding cuts, I noted back in July, it might be to force a reckoning inside schools about what’s gone wrong and how to fix it. Over in the New Yorker, Nicholas Lehmann updates the story with tales of the institutions that have simply settled, and those that are fighting back.

The premise of the great American universities today is a difficult one, to say the least: that they can be fantastically selective (but in a completely fair way), offer their students and faculty access to the most prestigious and well-rewarded precincts of American society, relentlessly increase their costs, assure the world that they are devoted to public service and social justice, and win the public’s grateful appreciation for being among our country’s most successful institutions. Trump, with his unerring talent for exploiting vulnerabilities in the liberal order, took full advantage of these contradictions and has caused enormous damage. It is going to be hard to undo.

Read more

2. Screen time

In some ways, the early adoption of television mirrored that of AI: initial skepticism followed by massive interest. But as writes in Works in Progress, for television at least, success was far from guaranteed. In fact, it faced some major structural obstacles AI doesn’t. Virginia explains how sports and kids became the medium’s saviors.

Amid widespread fears of ‘juvenile delinquency’, television promised to keep kids out of trouble. ‘There’s no juvenile delinquency here’, proclaimed an ad for TV sets, portraying a family seated around their living room console.

…In response, churches, Boys Clubs, youth centers, and YMCAs began buying their own sets. On a summer evening in 1948, a Methodist church outside Philadelphia attracted a crowd of 125, mostly teenagers, to watch a baseball game between the Philadelphia Phillies and the Cincinnati Reds. Pastors around the country reported that TV was drawing in the kids. ‘Our church set is used primarily for hockey games, prize fights, and other sports events’, said a Presbyterian minister in Chicago. ‘We’re sorry we didn’t think of it before’.

Read more

3. Claude whisperers

Speaking of skepticism: A majority of Americans may be neutral or skeptical on AI. But not coders. In the New York Times Magazine, Clive Thompson writes how AI solved one of coding’s main problems, drudgery. AI has transformed software development so dramatically, he writes, that most programmers now spend their days directing AI agents in plain English rather than writing code themselves, boosting productivity by anywhere from 10x to 100x. H/t Lux’s scientist-in-residence .

I looked at Ebert’s prompt file. It included a prompt telling the agents that any new code had to pass every single test before it got pushed into Hyperspell’s real-world product. One such test for Python code, called a pytest, had its own specific prompt that caught my eye: “Pushing code that fails pytest is unacceptable and embarrassing.”

Embarrassing? Did that actually help, I wondered, telling the A.I. not to “embarrass” you? Ebert grinned sheepishly. He couldn’t prove it, but prompts like that seem to have slightly improved Claude’s performance.

Read more

4. AI’s victims

I’ve long worried that the rise of AI may be pushing India toward a dangerous precipice. In Hyperdimensional, adds to the alarm. At an AI summit in Delhi, he found that India, among others, seems to be ignoring the likelihood of near-term superintelligence, instead focusing on smaller, more accessible AI models. He argues this is partly denial, partly a pushback against supposed American technological imperialism — but that this stance will ultimately leave countries dangerously unprepared for what’s coming.

Once countries have accepted the likely reality we occupy, a range of positive options become available. First and foremost, many Middle Powers and developing countries can bet that they have greater institutional flexibility than the more rigid U.S. Indeed, betting on continued American institutional sclerosis seems much safer than betting against deep learning. They can build new institutions, and reimagine existing ones, using the many new things frontier AI systems make possible (of which we have scratched the surface).

Read more

5. Bloom scroll

Finally some good news. The environment is recovering, Britain’s inactivity crisis is receding and Western power is underestimated. So why don’t we hear about it? According to in The Update, the negativity bias is strong with these ones.

Read more

From Riskgaming

From Lux Capital

Nominal, the startup aimed at making defense, space, energy and automotive manufacturing more efficient through advanced data analytics and testing, has achieved unicorn status. An $80 million investment round, including Lux, recently put the startup’s value at $1 billion.

“We wrote Nominal's first check,” Josh Wolfe noted in a post about the deal. “Since then? 7x ARR growth year-over-year. Four of the five largest defense contractors on the platform.” He continued:

When the engineers building the most technically advanced hardware systems in the world converge on the same platform, you pay attention. And when AI starts to actually WORK for hardware, not as a cheap “digital engineering” demo, not as a “digital twin” gimmick, but as a tool that earns trust by being right, it will be because platforms like Nominal built the data layer underneath it. You can’t have AI for machines without making everything about those machines machine-readable first.

In other news, Lux held its Health X Intelligence forum this week, including a panel on women’s health with Angela Dao of Maven, Bailey Renger of BeSound, and Melissa Teran of Alife. Fortune has a nice writeup of the discussion.

Finally, The Information reported on a memo from Josh to founders suggesting they start preparing for a bumpy road ahead. Looking to bonds, tariffs, infrastructure, and markets, he noted “signals suggest something is off.” From The Information: “he is concerned about ‘the bubble of AI, which people are all afraid to talk about publicly.’ He said many people in the VC industry ‘privately harbor these very vocal doubts’ but keep them quiet because of the upbeat culture.”

From around the web

1. Reading room

I’ll start off this week with an essay in Aeon dismissing all the doomsaying about technology and the decline of literacy. Human cognition is evolving rather than declining, Carlo Iacono writes, with people engaging ideas across multiple modes — text, audio, video — rather than just one — books. The solution isn’t nostalgia as much as making meaningful, multimodal thinking accessible for everyone.

Reading worked so well for so long not because text is magic, but because books came with built-in boundaries. They end. Pages stay still. Libraries provide quiet. These weren’t features of literacy itself but of the habitats where literacy lived. We need to rebuild those habitats for a world where meaning travels through many channels at once.

Read more

2. AI mode

Coming at this issue from a slightly different angle, a new Wharton paper offers an addition to Daniel Kahneman’s model of fast thinking (system 1) and slow thinking (system 2): AI thinking (system 3). Through three experiments, the authors demonstrate “cognitive surrender,” where people routinely defer to AI outputs with little critical scrutiny. H/t Katie Salam

Tri-System Theory is not a warning about AI’s dangers but a recognition of System 3’s psychological presence. We do not merely use AI; we think with it. In doing so, we must ask new questions: What happens when our judgments are shaped by minds not our own? What becomes of intuition and effort when a generative, artificial partner stands ready to answer? How do we preserve agency, reflection, and autonomy in a world where users engage in cognitive surrender? We offer Tri-System Theory as a conceptual foundation for understanding these challenges. It is a theory for an age of human-AI algorithmic cognition, and for the decision-makers, researchers, and designers shaping that future.

Read more

3. Tunnel vision

Remember Elon Musk’s hyperloop? Only vaguely? That’s by design. Laurence liked Matt Ribel’s look at what went wrong in the gambit to bring us revolutionary high-speed underground travel — an effort that had all but collapsed by 2021 due to technical incompetence, regulatory hurdles, and a lack of serious planning. For Ribel, this saga reflects a broader cultural tendency to embrace flashy, billionaire-led tech promises over the hard, collective work that real infrastructure requires.

Soon after, the Boring Company gathered reporters in California for a highly anticipated demo. But there was no high-speed, self-driving bus. Instead, a Tesla Model X SUV carried passengers through a short tunnel, sometimes approaching 50 miles an hour and ricocheting so violently that one journalist fell ill. Musk’s new-new vision was just a car, moving slower than it would aboveground. “I thought it was epic,” he told reporters. “It was an epiphany.”

Read more

For more on Musk’s foibles, check out my interview with Christian Davenport, author of the book, Rocket Dreams: Musk, Bezos, and the Inside Story of the New, Trillion-Dollar Space Race.

Read more

4. Lost in translation

We don’t have speedy transportation, but we do have AI that can beat us at chess. And science is not far behind. On that score, scientist-in-residence Sam Arbesman recommends ’ take on the “legibility problem”: What happens when AI makes discoveries that are too complex or conceptually foreign for humans to understand — and use.

If AI science does achieve superhuman performance, and if AI systems begin forming their own research communities around concepts that mutate faster than we can track, then the work of human scientists will shift from that of creation to that of excavation. How much knowledge we will need in order to act on what AI discovers remains an open question. But some degree of legibility will be essential, as discoveries that cannot be interpreted cannot be deployed.

Read more

5. Card trick

Finally, Sam directs you to a fascinating video about the life and death of HyperCard, which came before the web and led to the best-selling computer game of all time — then disappeared.

From Riskgaming

Chatham House

The Riskgaming team was in London this week hosting our European space sovereignty scenario Dead Reckoning during Chatham House’s annual security and defense conference. We’ll have more in the next few weeks to share, but what a distinguished group of policymakers! We’re very thankful that the think tank made official space for Riskgaming amidst the very busy programming.

On Citrini’s death of white collars

The viral research note of 2026 so far was Citrini Research’s prognostication of the complete collapse of America’s economy in 2028 thanks to a proliferation of intelligent bots that force hundreds of thousands of well-compensated, white-collar employees to line up at the unemployment office.

It’s thought-provoking and worth a read, although I am deeply skeptical of the trajectory it presents (Citadel seems to have the most popular counter-argument). Some random thoughts:

• Macro Feels Correct, Micro Feels Wrong: I agree with the note that AI will heavily disrupt middle managers and “middlemen” more broadly. Despite AI labor theorists’ focus on professionals and expert knowledge (think radiologists), the true productivity savings of AI is in cutting through the communications overhead within organizations and flattening out hierarchies. Nevertheless, all of the individual micro examples of such productivity seem wrong. For instance, credit card interchange fees. Citrini argues that they will head to zero as agents optimize for the lowest transaction costs. That would assume America follows Europe, which it hasn’t. Why? Because Americans love points. They’re aspirational, and agents in my view will actually increase the usage of points and thus interchange fees as they seek to get the best deal for their users.

• Missing Politics: The report doesn’t address any political blowback to the upper-middle class suffering from massive unemployment. Congress and the executive branch may not be the most responsive political bodies around, but they are truly responsive to this voting bloc. If widespread fears of job loss spread among elite Americans, these institutions would act, and act posthaste. That system dynamic must be factored in to make any analysis plausible.

• India is F’d: The report bleakly and briefly notes that India will suffer tremendously in the years ahead as outsourced IT services (representing hundreds of billions of dollars of exports) are replaced by agentic AI. I completely agree, and the global development community has yet to recognize the scale of the crisis. The same can be said of the Philippines and myriad other countries.

• Watch SaaS Private Credit: Citrini foresees a crash in the SaaS market given high debt leverage ratios against ARR underwriting and the coming substitution of AI. I am intrigued by this scenario and I am analytically neutral. Nevertheless, it’s hard for me to believe that enterprises, which generally subscribe to 100+ SaaS products simultaneously, will magically vibecode them all in one shot. This isn’t a one-year story but the narrative of the next decade — and possibly even more.

• Income Taxes May be America’s Achilles’ Heel: The note highlights the fiscal weakness of the United States given that income taxes could plunge with agentic AI. Perhaps hyperbole, but I do think the argument merits serious consideration as traditional purchases of goods and services (fully legible and taxable by Uncle Sam) transform into agentic work by AI (mostly illegible and untaxable). The U.S. deficit is catastrophically high, and any shrinkage of the tax base could have massive repercussions and lead to a spiraling doom loop.

From around the web

1. Bet the House

This week, Laurence recommends a profile of Alan Cole in the Wall Street Journal. Alan, a 37-year-old tax economist, bet his life savings on a Kalshi prediction market wager that, in 2025, U.S. federal spending would continue to rise, contrary to DOGE’s promises of dramatic cuts. It turns out he was right. Readers (and bettors) looking to avoid prediction markets’ adversarial ambiguity problem should take a look at how he structured his wagers.

The U.S. government has been described as an insurance company with an army. Now, with federal debt nearing 100% of gross domestic product, it’s an insurance company with an army and a giant mortgage.

The forces driving spending ever upward—inflation, an aging population, healthcare costs and interest payments—can’t change quickly.

Read more

2. Exposure therapy

Scientist-in-residence Sam Arbesman, meanwhile, was struck by the results of a new study written up in New Scientist. Researchers, it seems, have discovered that the blood of pediatricians — who are constantly exposed to childhood respiratory viruses — contains highly potent antibodies against RSV and other viruses. In lab and animal tests, three antibodies derived from their immune cells outperformed existing approved treatments by up to 25x. Maybe you should let your doctor cough on you at your next visit.

It is common lore among paediatricians that they become increasingly resilient to respiratory viruses over the course of their careers, says Trent Calcutt at Port Macquarie Base Hospital in Australia. “I’ve been working with paediatric patients for 10 years now and in my first few years, I would probably get two to three notable respiratory illnesses per year, and now I can go a year without getting any.”

Read more

3. Dam politics

Water has become a bit of a theme for us at Riskgaming. From water infrastructure security to floods to chip fabs, we’ve been fascinated by the way the conversation about our most basic resource has been changing in the last few years. To that end, ’s recent piece in Works in Progress stands out. Judge argues that, since the Clean Water Act of 1972, federal regulators have steadily ratcheted up water mandates that far exceed what local governments consider necessary, driving up household water costs and stifling urban growth. Water regulation, then, should be returned to local and state governments, who are better positioned to balance costs and benefits for their own communities.

The supposed water wars can largely be solved by redirecting water from low-value irrigation to high-value urban uses. The cost of water is much higher for urban users than irrigators, sometimes by orders of magnitude. But trades are not happening. Right now, because of federal mandates, utility pricing structures, insufficient infrastructure, and a general belief that water is scarce and should therefore be squirreled away and not used, many utilities don’t want more water. Instead they are trying to drive up costs and limit use. As the San Antonio Water System says, ‘it’s important we do everything we can to decrease our water consumption’. Unlike most services or companies, which see their job as meeting consumer demand, utilities see their job as limiting it.

Read more

4. Sili-con job

Finally, and as I noted above, the Citrini Research report on AI and joblessness really took off this week. For those curious about the hurdles still facing the automation of human jobs, ’s response to Citrini offers a useful roadmap. He writes that true AI replacement requires clearing several hurdles — technical capability, consistency, regulatory approval, and human trust — and most industries are still in the early game on each.

I can hit a perfect golf shot now and then, but that doesn’t put me on the PGA Tour. What separates amateurs from professionals is consistency. AI can produce impressive outputs, but replacing humans requires sustained, reliable performance. Today’s systems are non deterministic, meaning the same prompt can produce different results because of the underlying statistical nature of how they operate. Meanwhile, AI’s capacity for long-term memory and continuous learning are still evolving, and both are critical to LLMs being consistent. Consistency should be seen as a central factor to whether AI can truly take over a role in a given domain.

Read more

From Riskgaming

Yes, I know there hasn’t been a Riskgaming dispatch this week — one of our first misses in months. Heavy travel + big plans = bad writing schedule. More to come!

From Lux Capital

Terra Industries raised $22 million in a capital round led by Lux. The Nigerian startup is building autonomous security systems including drones, sentry towers and unmanned ground vehicles.

Financial Times reporting on AI finds that Wall Street investors have started to worry about automation’s impact on white-collar jobs outside the tech sector leading to sell-offs in the financial services industry, trucking and beyond. “It feels like a mob with bats looking for the next hit,” our own Peter Hébert is quoted as saying. “It’s indiscriminate.”

Finally, wrote about memory chip stocks in Key Context, featuring Josh Wolfe.

Smart investors like Lux Capital’s Josh Wolfe and Atreides Management’s Gavin Baker have been adamantly bullish - and right - on memory stocks for a while. They called out how memory makers aren’t just producing commodity chips anymore, the dynamic random-access memory (DRAM) for computers and consumer electronics. They have also become key suppliers of high-bandwidth memory (HBM) for AI servers, which are far more advanced and complex, dare I say, proprietary.

From around the web

1. Not so grand

One of my favorite articles of the last few months. For all my IR people out there: Think you know grand strategy? Check out the latest on the History Does You Substack. As it turns out, “grand strategy” is largely a myth — an intellectually prestigious but vague concept that attempts to bring retrospective coherence to what is, in practice, the very fragmented and improvised process of policymaking.

The further one moves from seminar rooms and closer to institutions, budgets, bureaucracies, and political incentives, the harder it becomes to sustain the belief that coherence is the natural state of affairs. What had once appeared as strategic design began to look like post hoc storytelling; what had seemed like foresight revealed itself as survival filtered through hindsight. Grand strategy does not fail because it was poorly taught or insufficiently refined, but because it asked the wrong questions in the first place.

Read more

2. Flight control

Some things are more strategically coherent than politics, apparently including efforts to eradicate mosquitos. Lux’s David Yang flagged a new study in the New England Journal of Medicine about a trial in Singapore, in which researchers demonstrated that infecting male mosquitoes with Wolbachia bacteria could reduce mosquito populations. Dengue infection rates also dramatically fell.

Read more

3. Left to chance

This week, Laurence Pevsner liked a new piece by , who argues that the left has tried so hard to dismiss AI as a scam (‘fancy autocomplete’) that it has ceded AI regulation to the right. This dismissal, Dan says, stems from a mix of academic philosophy, tech industry skepticism, and ideological echo chambers.

Still, despite the relative alertness from political quarters, it’s hard to avoid the impression that the right is more alert, both to AI’s opportunity and its danger. That doesn’t mean they are masters of wise AI policy. Both the accelerationists and the industrialists influential in the current administration show it is alarmingly often the opposite. It simply means that, between them and the Steve Bannon anti-tech wing, more or less the entirety of the movement agrees AI is not a fake technology.

Read more

4. Self-starter

Finally, Tess van Stekelenburg directs our attention to new work on the origins of life. Researchers have discovered a remarkably short RNA enzyme capable of copying both itself and its complementary strand. This finding suggests the start of life may not have required large, complex molecules and lowers the bar for how self-replication could have spontaneously emerged on early Earth.

‘It substantially lowers the bar that needs to be achieved non-enzymatically before ribozyme-catalysed replication kicks in,’ says Gianni. ‘If this is the path needed for life to emerge, this in turn affects the overall probability of life spontaneously emerging from pure chemistry, which we can estimate to be higher than before.’

Read more

From Riskgaming

From around the web

1. Power bank

This week, we’re dusting the snow off our shoulders and talking about power: who has it, why, and what they’re doing with it. To kick off, I’m recommending ’s fantastic essay on Robert Caro’s biography of LBJ. Henrik shows how true political operators view power less through formal structures (offices, titles) and more as an asset to be extracted from countless small and often overlooked sources (favors, relationships) that can grow into massive leverage over time.

Since he sees the world always in terms of tools that let him bend people to his will, Johnson becomes a true connoisseur of leverage. He notices potential sources of power in places where I’d never think there was any. For example, he seems to realize, sometime in his twenties, that knowing how to get someone else a job gives you purchase over them. So, while still working as a teacher, he begins collecting jobs. When he quits a position, he does it in such a way that he can pass the job on to one of his friends. And when they quit, he makes sure the job goes to another person he wants to control—often by orchestrating so that the person who has the job quits at the worst possible moment for the employer, and the person Johnson wants to get the job comes in the same day with their CV and explains how they can take over and solve the problem.

Read more

2. Moral machine

Since it is 2026, the obvious next question: What about AI? Does it have power? Agency? Anthropic seems to think so (or at least worries that it might have feelings). Lux scientist-in-residence Sam Arbesman liked Benj Edwards’ piece in Ars Technica examining Anthropic’s new 30,000-word “Claude Constitution.” The curious document treats Claude as a potentially conscious entity deserving moral consideration, and it is unclear whether the company’s move toward anthropomorphism reflects genuine philosophical uncertainty, strategic marketing, or both.

If you want a model to behave as though it has moral standing, it may help to publicly and consistently treat it like it does. And once you’ve publicly committed to that framing, changing it would have consequences. If Anthropic suddenly declared, “We’re confident Claude isn’t conscious; we just found the framing useful,” a Claude trained on that new context might behave differently. Once established, the framing becomes self-reinforcing.

Read more

3. Spectacle state

To more human concerns: Laurence found a fascinating new index that synthesizes multiple theoretical frameworks from comparative politics and political sociology to estimate authoritarian consolidation in the United States. The index characterizes the week of February 2 as a “Caesarist” moment for the country, where a government is ruling through spectacle rather than genuine consent. As analogues, it points to Taiwan and South Korea during those countries’ democratic transitions, and Argentina post-junta.

This is a dangerous moment precisely because it combines high authoritarian risk with low consolidation capacity, creating unpredictable escalation as the regime faces mounting resistance it cannot institutionally overcome.

Read more

4. Critical mass

Finally, statistical power! Lux’s David Yang shared a new study on Epstein-Barr. Almost everyone gets infected with EBV at some point. The big mystery is why some people never even notice they have it, while others get seriously ill. (At the worst end, EBV-related cancers kill 130,000–200,000 people each year.) Scientists have long suspected that differences in people’s genes partly explain why outcomes vary so much from person to person. But until now, studies trying to find these genetic links have been too small to get clear answers—which is why the researchers for this study found and validated a new method to study the genomes of a much larger group of people.

In 1964, Anthony Epstein, Yvonne Barr and Burt Achong observed actively replicating viral particles from Burkitt lymphoma, discovering the virus that now bears their names: the Epstein–Barr virus (EBV). EBV was subsequently recognized as the first known human oncogenic virus, the cause of infectious mononucleosis, and an agent in developing and exacerbating multiple autoimmune diseases. Despite these wide-ranging pathogenic roles, EBV infection is nearly ubiquitous, infecting >90% of adults worldwide, with most individuals remaining asymptomatic.

Read more

From Riskgaming

From Lux Capital

On Thursday, Palo Alto Networks, the cybersecurity firm, acquired Lux portfolio company Chronosphere, a platform built to give customers more control over, insight into and security for their data. Meanwhile, Fiddler AI, the Lux company focused on allowing developers and users to better understand, analyze, validate and monitor their AI systems at scale, raised $30 million in Series C funding. Another company, Astranis, announced it has signed a 9-figure deal with Oman to build the kingdom its first dedicated communications satellite.

Finally, Applied Intuition, hosted Jamie Dimon for a lecture on why companies fail. Their summary of the event is well worth the read.

From around the web

1. Intelligent design

Everyone is Claude Coding these days. Whether you’ve been trying it out or not, “10 things I learned from burning myself out with AI coding agents,” recommended by our scientist-in-residence, Sam Arbesman, is worth a read. Benj Edwards explains what they’re good at, what they’re not good at and why they’ll probably end up creating more work for humans rather than making us obsolete.

Due to what might poetically be called “preconceived notions” baked into a coding model’s neural network (more technically, statistical semantic associations), it can be difficult to get AI agents to create truly novel things, even if you carefully spell out what you want.

For example, I spent four days trying to get Claude Code to create an Atari 800 version of my HTML game Violent Checkers, but it had trouble because in the game’s design, the squares on the checkerboard don’t matter beyond their starting positions. No matter how many times I told the agent (and made notes in my Claude project files), it would come back to trying to center the pieces to the squares, snap them within squares, or use the squares as a logical basis of the game’s calculations.

Read more

2. Syn city

In a similar vein, I loved scrolling through Isometric NYC, a project from software engineer Andy Coenen. Andy used AI coding agents and image models to create an isometric pixel-art map of the city. He found that the coding agents were transformative for building software tools, but image models struggled with consistency and couldn’t reliably assess their own mistakes. Whatever work the process took, the end result is beautiful.

Explore

3. Easy as pi

Now for something completely analog: Sam also liked this charming argument for cooking with a slide-rule in Entropic Thoughts. The piece is also a sneaky math lesson; as the author notes “this is not so much about cooking as it is about having fun with maths in an almost useful way.”

Kitchen work is all about proportions: sometimes the recipe is for four servings but you need six; maybe the recipe calls for 80 g of butter but you only have 57 g, so you have to adjust the other ingredients to match.

We could use an electronic calculator to figure out the rescaled amounts, but a slide rule makes it so much easier.

Read more

4. Odds on

Earlier this week, I posted an interview with Dustin Gouker about sports betting and the broader implications of prediction markets for politics and beyond. Over in Harpers, Jasper Craven gets into the topic from a different angle: by placing wagers himself. Four years after New York legalized sports gambling, he’s spent $18,000 and explored the ins and outs of Las Vegas’ de minimis gambling addiction efforts.

As I held my head in my hands, Ted told me that he far prefers watching sports without money on the line. He no longer frets over the minutiae of passing yards, and can simply enjoy the pure, invigorating thrill of witnessing athletic excellence. As we left the stadium, I was feeling eczema-level itchy to recoup my losses, and Ted passed along a helpful tip given to him by a former sponsor. He said that if an addict ever finds himself in a casino, he should ignore the buzzy slot machines and focus instead on the faces of the people playing them.

Read more

5. Cleared for takeoff

Finally, new Lux-er Hank Anderson directs you to a series of policy memos on industrialization from the Knudsen Industrial Strategy Fellowship. In one, Divya Nagaraj proposes a National Autonomy Certification Framework (NACF) to address the gap in the United States between innovation and deployment of new technologies, especially in autonomous systems like drones and self-driving vehicles.

The U.S. is the clear leader in early-stage autonomy research, but this leadership stalls when systems move from controlled lab or synthetic environments to real-world deployment. This “deployment gap” threatens vital sectors like logistics, transportation, and manufacturing. The friction is visible across domains: from flying drones beyond visual line of sight (BVLOS) to fielding unmanned maritime systems alongside manned vessels.

Read more

From Riskgaming

From around the web

1. Same game

Earlier this month, I wrote about how to succeed as a straddle power: Rather than increasing strategic maneuverability, blunt decisions often transform a complex game of negotiations into a linear tug-of-war. It turns out, I may have been channeling a 1500s report from Daniello de’ Ludovisi, a Venetian diplomatic envoy, assessing how Venice might maintain careful strategic neutrality between Emperor and Sultan.

Now let your most excellent lordships consider how much that military is thereby becoming disordered and unnerved, and that state weakened; and although I am certain that this is known to Ibrahim (who is of good intellect, but not of such virtue as to be able to remedy so many inconveniences), yet the love he has for himself, much greater than that which he has for his lord, makes him prefer to be alone in a dominion that is not very strong, compared to what it could become through diligence, than to be one of many companions in a firmer and stronger one. From this, however, comes great benefit to your Serenity; for in order to remedy so many disorders, he applies all his efforts to maintain good relations with you, as I have been well able to judge, knowing that losing your friendship would perhaps, on some occasion, be to the detriment and ruin of that state.

Read more

2. Deep freeze

As in the 16th century, part of the reason the playing field is opening up for straddle powers is the shifting strengths and dynamics among the players at the top. As the United States faces down China (and Venezuela and Greenland?), it is also contending with headwinds at home. Laurence was impressed by Joseph Politano’s look at employment trends over the last year. The bottom line is that the U.S. job market has stalled dramatically, hitting young workers and low-income earners the hardest.

Read more

3. Hard pass

Maybe the weird job market is why so many influencers are trying to become boxers. In a piece in New Lines Magazine, Tam Hussein bemoans how Jake Paul, Andrew Tate and the like have turned boxing matches from displays of genuine mastery to spectacle entertainment. Their ability to headline fights against deeply talented world champions is an object lesson in the way we’ve come to value fame over hard work.

Once again, the rules bend for him. That, then, is my problem with Paul. By sharing the stage with Joshua, his presence teaches this generation that mastery doesn’t matter — that hard work, patience and time count for nothing.

I saw this exact thing over the holidays, when I attended a poetry reading by a friend. The first person read a “poem” about the escapades of her two dogs; she admitted she had written it using ChatGPT. It was witty and fun, I suppose. But what was she expecting — applause?

Read more

4. Just works

Another paean to hard work: Scientist-in-residence Sam Arbesman recommends Anil Dash’s “How Markdown Took Over the World.” You might think you don’t use the plain-text formatting system, but you probably do. Its the backbone of developer tools, messaging apps and AI systems alike, and it quietly succeeded because it was designed to solve a real problem, was released for free and arrived at the perfect moment.

It’s important for everyone to know that the Internet, and the tech industry, don’t run without the generosity and genius of regular people. It is not just billion-dollar checks and Silicon Valley boardrooms that enable creativity over years, decades, or generations — it’s often a guy with a day job who just gives a damn about doing something right, sweating the details and assuming that if he cares enough about what he makes then others will too. The majority of the technical infrastructure of the Internet was created in this way. For free, often by people in academia, or as part of their regular work, with no promise of some big payday or getting a ton of credit.

Read more

5. Pattern break

Finally, if you need to reset your brain, try Tiled.art, a website Sam likes dedicated to tessellations. You can make your own or check out the gallery.

Check it out

From Riskgaming

From Lux Capital

OpenAI and Anthropic have both recently introduced AI for healthcare products. In a piece on the rollout and what it means for the sector, Alex Konrad at Upstarts quotes our own Deena Shakir:

The headline isn’t that healthcare is big, it’s that the AI labs are finally treating it as unavoidable. Health already is one of the heaviest, stickiest use cases for large models, especially on the consumer side. What’s new is the seriousness on the enterprise side.

The partnerships with health systems and pharma signal a recognition that healthcare doesn’t change through standalone tools. It changes when technology shows up inside existing institutions and earns the right to stay.

Read more

From around the web

1. Dead end?

Iran is obviously on the brain this week as protests around the country — and the crackdown against them — swell. It’s hard to know what will happen next, but according to one anonymous writer in Iran: not much. Writing for Persuasion, the author points out a few core impediments to change: the regime’s opaque power structure, the resistance’s lack of a leader, and the appeal to the rest of the world of a the status quo.

The Islamic Republic is not a pyramid of power; it’s a maze. Every corridor looks like an exit until it loops back to the same center. The design isn’t accidental; it’s survival architecture.

Read more

2. Strategic optimism

Between Iran, Venezuela and Greenland, it’s been hard to know where to focus this week. In the interest of not losing sight of the other major geopolitical theme of the year, though, we’re recommending a few stories on China. First, for the AI crowd: At the recent AGI summit in Beijing, Chinese AI leaders debated whether and how soon China could lead global AI development. (Opinions ranged from cautious optimism to strong optimism.) Of note was the riff below from AI exec Tang Jie, who recently led his company Zhipu AI to a successful IPO.

Innovation often erupts when we pour in resources but stop seeing gains.

Scaling still brings value—you can go from 10TB of data to 30TB or 100TB. But we have to ask: Is it worth it? What’s the return on that 1 or 2 billion dollars? If the improvement is tiny, the math doesn’t work. Same goes for re-training the foundation and RL loops every time—it’s just not efficient.

What we need is a new paradigm to measure return: Intelligence Efficiency.

If your goal is to increase the upper limit of intelligence, brute-force scaling is the dumbest way to do it. The real challenge is: how do we achieve the same gains with less scaling?

That’s why I truly believe a paradigm shift is coming in 2026. And we’re working hard to make sure we’re the ones leading it.

Read more

3. Powered up

Last year, meanwhile, China installed over half of the world’s new global wind and solar capacity. In May alone, it even added enough to power all of Poland. Photographer Weimin Chu’s photographs in Yale E360 of the transformation helped me wrap my head around the scale of development.

Read more

4. Where babies come from

I enjoyed recent reporting in the Wall Street Journal about Chinese clients using America’s surrogacy industry in pursuit of eye-popping numbers of U.S.-born babies — sometimes dozens or more — in circumvention of China’s ban on domestic surrogacy. Not surprisingly, proposed restrictions on the U.S. side on such surrogacy have followed.

When Pellman called Xu Bo in for a confidential hearing in the summer of 2023, he never entered the courtroom, according to people who attended the hearing. The maker of fantasy videogames lived in China and appeared via video, speaking through an interpreter. He said he hoped to have 20 or so U.S.-born children through surrogacy—boys, because they’re superior to girls—to one day take over his business. Several of his kids were being raised by nannies in nearby Irvine as they awaited paperwork to travel to China. He hadn’t yet met them, he told the judge, because work had been busy. The market has grown so sophisticated, experts say, that at times Chinese parents have had U.S.-born children without stepping foot in the country. A thriving mini-industry of American surrogacy agencies, law firms, clinics, delivery agencies and nanny services—even to pick up the newborns from hospitals—has risen to accommodate the demand, permitting parents to ship their genetic material abroad and get a baby delivered back, at a cost of up to $200,000 per child.

Read more

5. Scarymandered

On the U.S. side, the vibes are even weirder. This week, our scientist-in-residence Sam Arbesman recommends Frank Jacobs’ look at our cultural fascination with post-apocalyptic maps, from the geography of “Station Eleven” to “The Last of Us.” A common theme, he writes, is balkanization — fragmentation into smaller hostile regions based on religion, ethnicity, or politics — which underscores our anxieties about the precariousness of the American experiment.

America is rehearsing its disaggregation in the safest possible arena — that of science fiction, the comments section of history. In the safety of post-apocalyptic fiction, America is fabricating a future that is nostalgic for the present. The message is that, even with all of its problems and divisions, today is valuable enough to want to preserve. But should the worst happen, then even on the other side of a possible future apocalypse, there is still hope. Why else would anyone make a map of it?

Read more

6. Command control

If the United States does break apart, I’m guessing Larry and David Ellison will want to rule whatever group is based out of Hollywood. Laurence nicely captures the father-son dynamic analyzed by Reeves Wiedeman in New York Magazine: Great long piece, but can be summed up by one Steve Jobs anecdote in it:

Steve Jobs told one of Larry’s biographers that he was impressed that his friend was even willing to go up in a plane with teenage David at the controls. “He’s risking his life with his son,” Jobs said. But when Larry heard this comment, he said Jobs was forgetting one thing. “The fact of the matter is that if David does something stupid, I’ll grab the stick,” Larry said.

Read more

7. Genius genetics

Before I go, I urge you to check out Richard Stone in Science. His tale of the search for Leonardo da Vinci’s DNA will take you from life in fifteenth-century Tuscany to the efforts of a cutting-edge, globe-spanning genealogical project. Science has nearly progressed enough to use touch DNA to authenticate disputed artworks or even discover the nature of the old masters’ genius — with major ramifications for today’s art markets.

Yet some of what made Leonardo unique seems rooted in biology. His extraordinary ability to capture subtle shifts of light and motion, for example, has long hinted at exceptional visual acuity. LDVP aspires to one day find genetic variants that could account for it, says Gonzalez-Juarbe, who works at the University of Maryland. “Our hope is to open a door to explaining what was so unique about the smartest guy in history.”

Read more

From Riskgaming

A special announcement from Lux Capital

We’re very fortunate to announce our new fund, Lux Ventures IX. With over $1.5 billion, it’s the largest fund in our history. While the number is new, the thesis is not: we’re going to continue to invest where we have since our founding at the cutting-edges of frontier science and technology. “We happen to be in the companies delivering this at a critical time,” Josh Wolfe noted to Bloomberg. “Many feel very confident that this will be a company-making next few years for some” startups.

As we wrote in our announcement:

Today entire sectors of aerospace, biotech, defense, industrials, transportation and beyond are being reinvented by a new generation of brilliantly ambitious, often irreverent, scientists and engineers. And while markets have changed and capital reshapes around fewer companies and larger checks, our focus on the craft remains constant.

Thank you to our pathbreaking founders, our enthusiastic limited partners and the entire Lux ecosystem who make this possible.

Read more

From around the web

1. What’s new?

Starting something new in 2026? Be warned. It might end up looking pretty similar to everything that came before. In a new post, our scientist-in-residence explains why: canalization. It’s an idea borrowed from biology that explains why lots of different genetic combinations end up producing similar traits despite variation, environmental forces and randomness. Sam argues the principle applies outside of biology, too. For example, new research organizations tend to converge toward familiar institutional forms despite attempts to create novel structures.

With organizations that end up looking university-like, it comes down to risk, or more precisely, risk aversion. Imagine you build a weird institution devoted to some odd interdisciplinary topic. Then, you hire people to work in it. These are often academic researchers with varying levels of exposure to academia’s incentives and requirements. Each hire then begins to think the following: “What if this organization crashes and burns? Or what if I’m simply not a good fit for it? I need to make sure I can get another job afterward.” And for the most part, “job afterward” means another academic job.

Read more

2. Think tank theater

For me, Sam’s piece raises the question of what all the new think tank formats sprouting up are really for — a question I also got into this week on the podcast. Coming at this issue from a slightly different perspective is an article Laurence found over on Brookings. Author Jeremy Shapiro’s surprisingly amusing narrative about the secret rituals of meetings between government officials and think tank experts shows how the blob uses “thinkers” to validate their preexisting choices rather than to learn.

[The thinker] clears his schedule of any conflicting brown bags on separatism in South Ossetia and, after a suitable interval to keep the government guessing as to his availability, replies that he might be able to squeeze it in to his schedule. Citizenship data and social security numbers are provided for security purposes, times are confirmed and ground rules are established in a multitude of emails with a seemingly never-ending array of staffers, all of whose titles include the word “special.” The thinker says nothing directly to his colleagues, but searches desperately for opportunities to obliquely allude to the meeting: “I’d love to come to your roundtable on uncovered interest rate parity, but I unfortunately have a meeting with the secretary of defense.”

Read more

3. Burn, baby, build

Speaking of validation. This week, Lux’s Shaq Vayda recommends a great new exploration of the economics of AI data centers in space from Andrew McCalip (of portfolio company Varda). The idea, although far less “obviously stupid” than Andrew says he first thought, is really only feasible under aggressive assumptions and if one company (ahem, SpaceX) decides to throw money at the problem basically for funsies.

I’ll go one step further and say the quiet part out loud: we should be actively goading more billionaires into spending on irrational, high-variance projects that might actually advance civilization. I feel genuine secondhand embarrassment watching people torch their fortunes on yachts and status cosplay. No one cares about your Loro Piana. If you’ve built an empire, the best possible use of it is to burn its capital like a torch and light up a corner of the future. Fund the ugly middle. Pay for the iteration loops. Build the cathedrals. This is how we advance civilization.

Read more

4. Turing taste

On to more earth-bound concerns. Laurence also liked Vauhini Vara’s latest in The New Yorker. She asks what happens if people end up liking AI-generated fiction more than the real deal. Based on my usage of these tools, it sounds far-fetched. But maybe not. Vauhini reports on a computer scientist who fine-tuned AI models on the complete works of 30 prominent authors and found that creative writing students preferred the AI-generated prose to human imitations in nearly two-thirds of cases.

Reading the authors’ original passages alongside the A.I. imitations, I was startled to find that I liked some of the imitations just as much. The A.I. version of Han’s scene, about the newborn’s death, struck me as trite in places. But, to me, the line about the mother’s chant was more surprising and exact than the original. I also spotted some good bits in an imitation of Junot Díaz. In “This Is How You Lose Her,” Díaz writes, “The one thing she warned you about, that she swore she would never forgive, was cheating. I’ll put a machete in you, she promised.” To my ear, the A.I. rendition was more rhythmic and economical: “She told you from the beginning that if you ever cheated on her she would chop your little pito off.”

Read more

5. Toy story

If you’re like me, this first week back after the holidays was a bear, so let’s ease on out with something fun. In March, Lego is launching “Smart Bricks” — bricks with tiny computers that communicate with NFC-equipped tiles and minifigures to add interactive sound, light, and motion effects to builds. Many on the Lux team have already added them to our Christmas lists for next year.

Feeling lo-fi? Perhaps PastaTube is a better fit.

From Riskgaming

From Lux Capital

Atavistik Bio, which focuses on the discovery and development of novel disease therapies, has raised $120 million in Series B funding from Lux and a syndicate of others. Another portfolio company, Medra, which programs robots with artificial intelligence to conduct biological experiments, has raised $52 million for an autonomous lab. Finally, Unconventional AI, developer of next-generation computers designed for AI and efficiency, has raised $475 million from Lux and others.

What you clicked on this year in Lux Recommends

As 2025 winds down, I wanted to close out the year with something a little different: a look back at the stories that most caught your attention.

In order below are the ten reads you clicked on more than any others this year. Some, I expected; others surprised me. Together, they paint an interesting picture of what mattered — or at least what intrigued — this year.

Consider it a collective reading list, a time capsule, or just a good excuse to revisit something you meant to finish. Either way, thanks for reading along this year. We’ll have a year-end roundup of Riskgaming early next week, and we’ll be back with more Lux Recommends in 2026.

1. Anxiety, now boarding

I’ve spent a decent amount of time in airports recently, and when I came across JA Westenberg’s “Boarding Group One,” I felt seen. She observes how airline boarding groups have become a perfect microcosm of our status-obsessed culture, where status anxiety has metastasized into every corner of modern life, from the cars we can’t afford to the wine we order to impress strangers, to competition in friendships and even parenting.

I don’t want to pretend I have a neat solution. I can’t tell you to renounce status and expect you to listen; if I’m being brutally honest, I wouldn’t follow that advice. We all like the way it feels to get called first and I include myself in that indictment. But I can tell you this: the way we let status rule our lives is making us miserable. It pits us against each other in a zero-sum game where even the winners can’t relax, because there’s always another rung, another group, another measure. If you’re in Group Two, you want to be in Group One. If you’re in Group One, you yearn for a private jet. There’s no end to it.

Read more

2. Lost in the flow

For Zelda Breath of the Wild and Tears of the Kingdom fans, the video below on the rivers of Hyrule may be the best 30 minutes of your week — heck, the best of your year.

3. CapEx Machina

In 2026, we’ll all still be sorting AI fact from fiction, but as a percent of GDP, AI CapEx has already surpassed spending during the dot-com bubble and the 2020s telecom boom. Using all this money for AI means it isn’t paying for other things, though, and in this article recommended by Lux’s Grace Isford, Paul Kedrosky investigates what’s losing out.

One of the abiding mysteries of the current political era is why the economy is, for the most part, not as worried as one might expect about tariffs, political uncertainty, capricious office-renovation-driven-Fed-chair rumored removals.

We now have a possible answer. In a sense, there is a massive private sector stimulus program underway in the U.S.. There is an AI datacenter spending program, one that is reallocating gobs of spending, as well as injecting even more.

Read more

4. Digital drift

If you’re in a contemplative move, revisit this beautiful meditation by Karl Ove Knausgård in Harpers on what we’ve lost by trading deep but circumscribed lives in the real world for shallow but vast digital existences.

Throughout history it’s been said that mathematics is the language of nature, but I couldn’t even grasp something as simple as the relationship between the language of mathematics and the laws of the physical world. If numbers were abstract and math was derived from fundamental axioms, it would be a closed system, and in that case how could it describe and calculate the most breathtakingly complex events in the natural world—which weren’t abstract at all?

Read more

5. Build, baby, build

Or if you just want to make something, Lux’s Shaq Vayda plugs LegoGPT, an AI tool to help you generate LEGO structures from text prompts.

Play around

6. Watt really happened

Earlier this year, initial analysis of Spain’s April blackout heaped the blame on renewables, especially wind and solar, because of their intermittency. But it’s worth checking out this post Laurence found from Avila VC that shows why renewables weren’t really the problem.

Renewables put pressure on systems to be more flexible and resilient, but the tradeoff is access to the most unlimited, cheapest, and cleanest form of energy we have today. Spain’s recent blackout wasn’t caused by renewables—it was a grid management failure. Meanwhile, smart countries like China are building flexible grids alongside massive amount of renewables and nuclear.

Read more

7. Where alpha’s at

Josh’s annual Lux LP letter is out (this was Q3), with new reflections on our origin story, some of our biggest mistakes, and where things are headed from here. I found this part especially useful:

You can find his annotated version of the letter on X.

Check it out

8. In the fold

Onward to food. Din Tai Fung, which debuted in the United States as a family-run suburban LA strip mall restaurant 25 years ago, is now America’s highest-earning restaurant chain. Under CEO Albert Yang, the family has driven growth by obsessively focusing on guest reviews, standardizing operations (every dumpling must have exactly 16 grams of filling, 5 grams of dough, and 18 folds), and refusing to franchise or go public to maintain quality control. If you get a chance to check it out, you should.

When the brothers began managing the American business, which operates separately from the international arm, the brand began focusing more on data and guest feedback, with Yang calling high guest ratings his “north star.” Yang, who earned a Cornell University degree in hotel administration in 2015, and his team obsessively monitor reviews on Yelp and Google, and corporate bonuses are tied to good reviews. In the restaurant industry, the most common complaints usually center around cold food or inattentive service. To address the former, Din Tai Fung increased cook times on certain dishes to ensure hotter temperatures. For the latter, he’s got a program of intense training. “It’s just nonstop—‘How do we improve the guest experience?’” says Yang as a pot of oolong tea arrives steaming hot.

Read more

9. Rules for fun

As we stare down the holidays, let’s talk about parties — specifically, what makes a good one. Luckily my good friend and long-time Riskgaming participant, Uri Bram, has some rules over at Atoms and Bits. The next time I put together something for Riskgaming, I’m keeping #20 in mind!

20) Let me repeat that: Parties are a public service, you’re doing people a favor by throwing them. Someone might meet their new best friend or future lover at your gathering. In the short term, lovely people may feel less lonely, and that’s thanks to you. In the long term, whole new children may ultimately exist in the world because you bothered to throw a party. Throwing parties is stressful for most people, but a great kindness to the community, so genuinely pat yourself on the back for doing this.

Read more

Uri also has a very fun game out. Person Do Thing would help any party along, and it is available digitally and as a card game.

10. Shell games

Finally, if your New Year’s resolution includes more lies, I would recommend a recent study on secrecy strategies from three researchers at Dartmouth. They track efforts by elites around the world to conceal their assets and identities and find three basic patterns.

Overall, our most counter-intuitive finding is that use of offshore finance by elites can be driven not only by poor governance in their home countries—like corruption and lack of civil liberties—but by good governance conditions, as well.

Read more

From Riskgaming

Truth Santa

This week, a dictatorial Santa redirected the elves of Santa’s Workshop from global gift giving to building up the North Pole into an AI superpower.

You really had to be there. You had to be there for the comedian joking about how “the North Pole has embraced late-stage capitalism and it is gorgeous. They’re going to start calling me Santa Close, because I am always closing.”

You had to be there for the Lina-Khan-o-meter that determined the North Pole’s antitrust policy (reasonably generous!)

You had to be there for the launch of $antaCoin, for Ballmer Lucky’s haircut and for the dozens of Truth Santa posts put up by our eggnog-drinking Santa.

And you had to be there as players searched the room for PornHub datasets in their naughty quest for AI supremacy.

This was the most ridiculous Riskgaming experience we have ever done. And it’s never happening again. So for those who couldn’t be there, we have posted a bit of the story and materials on an unlisted website that will live in perpetuity. In the meantime, have a great holiday!

Read Truth Santa

From around the web

1. Fine print

I hope you get some time to relax with a book over the holidays. If you do, try close reading — the practice of noticing small textual details and building arguments from them. I’ve been inspired by this article from Johanna Winant, a former West Virginia University professor, who describes how teaching her students to read this way helped them realize they could trust their own observations and make meaningful claims about their world.

Let’s start by looking at the text in front of us and pointing to something very small: a single word, or punctuation mark, or even something that’s not there, a gap. Tell me, I would ask as we sat down and opened our books, what is one detail that you noticed? What snagged you? Where were you surprised? At the beginning of every semester, my students would be confused by these questions. They were smart, hardworking young people, and they very much wanted to get the answer right. They pointed to themes, identified genres and symbols, and gestured toward historical contexts.

Okay, I would respond, but now point to a detail, one that’s really on the page and small enough to fit under your finger. I sat, and smiled a little, and waited with the conviction that I wouldn’t be disappointed. And then, reliably, every time, a transformation that seemed like magic.

Read more

2. Lower ed

If close reading is a skill universities should be teaching , most of them have lost their way. This week, Laurence flags a piece in the FT by a colleague of his at the Moynihan Center. Carlo Invernizzi Accetti argues elite U.S. universities made themselves vulnerable to the Trump administration’s political attacks by prioritizing research prestige over their civic mission of forming engaged citizens.

The political weather will change. Courts will (one hopes) curb excesses. New administrations will rewrite rules. But legitimacy does not come only from a friendlier federal government. It also comes from the daily experience of students who emerge as capable citizens, and from communities that can see the value those students create.

Read more

3. Elite loitering

Enough learning. Let’s move on to holiday travel and the airport lounge wars. If you haven’t read Zach Helfand in The New Yorker yet, you should. Zach took a week-long tour of airport lounges across New York so you don’t have to. From bare-bones Priority Pass offerings with mushy cheesecake to ultra-luxurious spaces with Danny Meyer restaurants, spa facials, and Porsche rides to the plane, choose your travel route this year carefully.

We used to spend a lot more time waiting than we do now. We waited for the mail, for the milkman, for the news, for a ship, for a sign, for the bread to rise, for the tide to ebb, for the cavalry, for good things to come. As people were always waiting for something, dedicating special areas in which to do so would’ve been ludicrous. In the sixteenth century, kings, Popes, Medicis, and other aristocrats began constructing rooms where courtiers would wait. In “The Antechamber: Toward a History of Waiting,” the historian Helmut Puff recounts that, when Mozart was twenty-one and seeking a patron, he complained in letters home about waiting in antechambers across Europe: an hour in Bavaria, another “whole hour” in France, a half hour in a frigid room of a duke and a duchess. When the duchess finally appeared, he told her, “I’d be only too happy to play something but that it was now impossible, as my fingers were numb with cold.” Waiting can make one feel needy, like a baby. The waiter waits because the waitee is too important to.

By contrast, waiters wait in the airport lounge because they are important.

Read more

4. Flight risk

Can’t travel yourself? Maybe send a rocket. Some amateurs are getting their projectiles nearly to space. Laurence liked Michael Harmon’s reporting from one club’s monthly launch. The Metra Rocket Club brings together amateur rocket enthusiasts — from retirees to university engineering students — at a New York farm to send off rockets so high-powered they need FAA authorization.

One by one, the rockets shot into the sky. “Deploy, deploy,” Mr. Marote intoned, speaking directly to his rocket’s parachute.

Necks craned skyward, the families cheered as each rocket’s parachute deployed as designed. The men would earn their certification as long as they could recover their rockets undamaged (which they did).

Read more

5. Size matters

Finally, our scientist-in-residence Sam Arbesman is here to surprise and delight you with a new visualization by Neal Agarwal, of Neal.fun. He maps the size of various life forms — from the humble strand of DNA (2.5 nanometers tall) to the Pando Clone, which looks like a forest, but is actually a single, 108-acre organism in Utah, composed of over 47,000 stems connected by a vast root system.

Check it out

From Riskgaming

From Lux Capital

Josh’s annual Lux LP Letter is out, with new reflections on our origin story, some of our biggest mistakes, and where things are headed from here. I found this part especially useful:

You can find his annotated version of the letter on X.

In other Lux news, Trial Library just raised a $10 million Series A to better match patients with clinical trials. Portfolio company Varda, which is building space factories to launch and return space-made products for terrestrial applications, was profiled in TechCrunch:

When Will Bruey talks about the future, the timelines are shorter than most might imagine. The Varda Space Industries CEO predicts that within 10 years, someone could stand at a landing site and watch multiple specialized spacecraft per night zooming toward Earth like shooting stars, each carrying pharmaceuticals manufactured in space. Within 15 to 20 years, he says, it will be cheaper to send a working-class employee to orbit for a month than to keep them on Earth.

Finally, check out Wired’s reporting on Fortell’s AI-powered hearing aid (formerly known as Chromatic). The portfolio company’s device is inspiring long waiting lists of people who want to try it out — and tears of joy when they do.

From around the web

1. Daylight savings

How’s your sleep? Rowan Jacobsen’s is horrible. In a new article for Harper’s, he chronicles his attempt to cure his chronic insomnia by spending a month in the Arizona desert living without artificial light after dark. The key insight: it’s not just avoiding light at night that matters, but maximizing the amplitude between bright days and dark nights — the intense daylight exposure our ancestors evolved with but modern indoor life has erased. Good luck with that this dark December.

It feels strangely radical to have the sunset as one’s entire agenda. It passes through a dozen stages, from fiery orange to astral lavender, a menagerie of photons that left the solar corona eight minutes ago and banked their way into me. All the while my light meter is slowly ticking down, from a few hundred lux to just dozens. Finally, the color is gone, except for some peaceful blue bands along the skyline. It seems clear that we’re built for this slow fade, rather than lights-out at eleven.

Read more

2. Shift happens

In more charmingly off-beat “self help,” our scientist-in-residence Sam Arbesman recommends Kurt Streeter’s “How to Fix a Typewriter and Your Life” in The New York Times. The story follows a burned-out facilities manager named Paul Lundy, who in 2015 read a newspaper article about 92-year-old typewriter repairman Bob Montgomery. Paul became his apprentice, eventually bought his business, and found a new purpose late in life.

He showed Lundy the right way. No anger. No frustration. Just quiet insistence that good enough was not good enough.

Sometimes Mr. Montgomery would partly disassemble a machine and leave it on Lundy’s bench — a test, a puzzle, a method of teaching as old as apprenticeship itself.

Read more

3. Grid expectations

If the United States wants to remain competitive with China in AI, it needs to do something new, say Jared Dunnmon, Avanika Narayan, Jon Saad-Falcon, and Chris Ré. While China is limited by export controls on advanced chips, the United States remains deeply constrained by power grid bottlenecks. The solution: maximizing “Gross Domestic Intelligence” — the total computational intelligence a nation can deploy — rather than just building more data centers. This will mean activating the millions of AI-capable laptops and phones already in American homes through hybrid local-cloud inference systems, which could boost usable AI capacity by 30–40% without building new infrastructure.

Maximizing deployed intelligence requires two simultaneous objectives: making computation more intelligent per unit of power (better models, better chips), and maximizing the power available to run that computation. Miss either dimension and you lose: efficient models without power sit idle; abundant power without efficient computation drives up costs with diminishing returns. In this framing, the absolute amount of intelligence a nation can deploy is the product of intelligence per watt and the amount of power usable for computation — what we call Gross Domestic Intelligence (GDI).

Read more

4. Green tech, red tape

Also harming U.S. competitiveness: over-regulation. Laurence liked Peter Reinhardt’s discussion of his travails trying to start two climate-tech hardware companies: Charm Industrial (carbon removal) and Revoy (electric trucking). Peter argues that more than half the cost of building each company went to regulatory burdens. He details how a 5.5-year wait just to determine which permitting pathway to use cost Charm $125M directly and $300M in public health costs from continued pollution.

Regulatory delays and bottlenecks have added millions of pounds of pollutants like PM2.5, NOₓ and CO₂ to our air from the continuation of business as usual, instead of the deployment of clean technologies from my two hardtech efforts alone. While CO₂ is a long-term climate issue, PM2.5 and NOₓ are immediate major drivers of asthma and excess morbidity. Both operations have high bipartisan appeal—and we’ve never been denied a permit—because we’re fundamentally cleaning up things that matter to everyone: dirty air, wildfires, orphaned oil wells. Revoy is also helping deflate the cost of long-haul freight. But none of that has made getting freedom to operate easy. For creative new technologies the default answer is “no” because there isn’t a clear path to permitting at all, and figuring out that path itself takes years — time that startups can’t afford to wait.

Read more

5. The thin red line

Laurence also suggests checking out the debate over the poverty line between and Jeremy Horpedahl. Michael argues America’s poverty line is catastrophically outdated, and that updating the formula with today’s spending ratios yields a real crisis threshold of around $140,000 for a family of four (rather than closer to $31,000). This means, he says, that the median household earning $80,000 is actually living in poverty.

When you run the net-income numbers, a family earning $100,000 is effectively in a worse monthly financial position than a family earning $40,000.

At $40,000, you are drowning, but the state gives you a life vest. At $100,000, you are drowning, but the state says you are a “high earner” and ties an anchor to your ankle called “Market Price.”

Jeremy retorts that Michael made three major errors: he underestimates how much income American families have; treats average spending as minimum spending, overstating costs by 30-40%; and ignores that the Census Bureau already has a better measure than the poverty line — the Supplemental Poverty Measure. The debate between the two is ongoing.

Read the original piece here

Read the first response

From Riskgaming

Doomberg

It’s not everyday I interview an animated bird on Zoom for a conference, but hey, that’s the life of a writer au capital venture in 2025. I recently interviewed Substack-extraordinaire Doomberg for the SOSV Climate Tech Summit, covering topics like the future of natural gas and whether climate change is, like, real real or just real, or just… you know, they say that “a bird in the hand is worth two in the bush,” so you be the judge of whether I got a handle on this interview.

From around the web

1. My two cents is now a nickel (inflation!)

Let’s start this week with an ode to the penny… Actually, I have nothing nice to say. You probably don’t either. But whether you supported the Americans for Common Cents (pro-penny thanks to pro-zinc forces) or the anti-forces of “it just makes sense” — the demise of the penny is an historic moment for America and the English language. A penny for your thoughts is becoming a quarter — and no one wants to pay that much for a thought (hence why the Riskgaming newsletter is free).

This week, our scientist-in-residence Sam Arbesman recommends Caity Weaver’s look at what will happen to the 300 billion circulating pennies after the government decided to abruptly stopped minting them. Surprise! There is no plan whatsoever. According to Caity, the government is essentially abandoning responsibility for the massive stockpile sitting in American homes and businesses, treating it as they always have: the public’s problem.

Most pennies produced by the U.S. Mint are given out as change but never spent; this creates an incessant demand for new pennies to replace them, so that cash transactions that necessitate pennies (i.e., any concluding with a sum whose final digit is 1, 2, 3, 4, 6, 7, 8 or 9) can be settled. Because these replacement pennies will themselves not be spent, they will need to be replaced with new pennies that will also not be spent, and so will have to be replaced with new pennies that will not be spent, which will have to be replaced by new pennies (that will not be spent, and so will have to be replaced). In other words, we keep minting pennies because no one uses the pennies we mint.

Read more

2. Cover stories

Sam also liked Dan Piepenbring’s interview with Lance Richardson, who recently published a biography of Peter Matthiessen, the man who co-founded The Paris Review in 1951 while working undercover for the CIA. Richardson explores whether the CIA ever directly funded the magazine and how Matthiessen grappled with shame over his intelligence work. The interview is a nice reminder of the ways in which American literary culture has — and hasn’t — changed.

The problem with Matthiessen’s cover soon became clear—the labor of a writer is pretty invisible to the outside world. It looks like we’re just sitting inside and not doing anything at all. Matthiessen’s handler told him he needed a visible profession. And one day in one of the cafés he runs into Harold “Doc” Humes, another American who was running a magazine called the Paris News Post, which he had acquired for six hundred bucks, because that was the trend among expats in postwar Paris. Everyone had a little magazine going in that time—there was Merlin, Points, Zero.

Humes was a real character, a bit of a loose cannon. He was wearing a cape when Matthiessen saw him at the café that day. He brought on Matthiessen as his fiction editor. But Matthiessen saw the Paris News Post as a lightweight endeavor. He suggested one day to Doc that they flick it off and make something better.

Read more

3. Devil in the details

Speaking of spies, Robert McMillan’s reporting on the effort to take down a North Korean hacker known as “Hades” has all the twists and turns of a thriller. This year, two former FBI cybersecurity researchers finally got him by posing as an employer and offering him a programming job. As it turns out, the hacker was now working as one of North Korea’s thousands of IT workers who assume fake identities to obtain legitimate work with Western corporations.

Last year, North Korea made as much as $800 million from its thousands of worldwide IT workers, who assume fake identities and obtain jobs with Western corporations, according to a report by 11 countries that are monitoring North Korea’s sanction violations. Workers can make as much as $100,000 a month, much of which is funneled back to support agencies such as the RGB [Reconnaissance General Bureau] spy agency.

Read more

4. Bad timing

AI has a time problem. Just check out artist Brian Moore’s “AI World Clocks” page. Every minute, nine different AI models are asked to generate a new analog clock showing the current time. The results have me puzzling over what, exactly, makes this exercise so complicated for our AI overlords — and gives me hope that humans may at least count on steady employment as railroad conductors for decades to come.

Check it out

5. Mind the gap

Finally, offers a new way to look at the Indian economy. The country’s per-capita income is around $2,400 for 2025, but this figure masks extreme fragmentation. In fact, the country really functions as nine distinct economies, with the top three urban cohorts (wealthy, affluent, and emerging) representing only 15% of households but commanding 68% of total income. This concentration means India’s consumption growth story centers on a small, rapidly expanding urban cluster whose spending patterns increasingly resemble middle-income Asian economies, while the majority remains stuck in lower income bands due to a lack of job creation.

The behaviour and expansion of these cohorts have a direct implication for India’s consumption story. For consumer brands, the practical addressable market over the next decade is less “1 billion Indians” than a set of dense urban corridors whose income pools, in CLSA’s mapping, already resemble entire economies such as Brazil, South Korea, Switzerland or Sweden rather than the national average.

Read more

From Riskgaming

Shameless Self-Promotion

The Special Competitive Studies Project has a nice writeup on the recent collaboration with us on a run-through of Southwest Silicon. Among other takeaways, SCSP noted in their newsletter that:

Tradeoffs aren’t failures, but the price of progress. Every major technology investment comes with competing demands for scarce resources, like water, energy, land, and even capital. The game underscored not just that tradeoffs exist, but how quickly and fairly we can negotiate them. As America sets its sights on an era of reindustrialization, these state and local-level negotiations will be the true front lines of building the next-generation of technology infrastructure – from fabs to data centers to fusion machines and beyond.

From Lux Capital

Majestic Labs, a Lux portfolio startup reimagining AI infrastructure to reduce cloud companies’ spend on data centers, announced this week that it has raised $100 million. Meanwhile, Israel-based cybersecurity company Tenzai came out of stealth. The company is aiming to use AI to help customers find vulnerabilities in their critical systems.

Bilal Abu-Ghazaleh, formerly of Scale AI, has raised $9 million in seed capital, including from Lux, for his new firm, 1001 AI, which seeks to build AI-native operating systems for critical industries. And the New York Times published a major report on successes in the field of xenotransplantation, including those led by Lux’s eGenesis, which next year will begin a “trial of kidneys transplanted from pigs that have undergone 69 gene edits.”

Finally, if you missed the Lux AI Summit, check out the videos from the event on Lux’s YouTube channel, including this session with Scott Wu of Cognition. Scott discusses the future of software engineering, predicting major advancements in autonomous AI agents over the next 12 months.

From around the web

1. Red teaming

I’m starting off this week with a look at a new (non-Riskgaming) war game: Syracuse University recently had experts role-play as Chinese military planners considering Taiwan invasion scenarios. By forcing players to take on the Chinese perspective rather than the American one, the exercise exposed a blind spot in U.S. military planning — we neglect China’s incentive to pursue coercive strategies that exploit American political hesitancy to enter a war where U.S. forces weren’t attacked first. This is exactly what wargaming is designed to do!

The exercise forced participants to confront an uncomfortable truth about China’s military capabilities. Despite decades of modernization, the People’s Liberation Army has not fought a major conflict since 1979. It has never conducted an amphibious assault on a major scale. Its logistics remain untested. Its command structure is riddled with political interference. In contrast to most wargames that portray the Chinese military as a competent machine operating at maximum efficiency, the perspective from Beijing is likely more sobering.

These limitations don’t make China weak — they make it cautious. Why attempt a Normandy-style invasion when missile strikes and economic strangulation might achieve the same goal? Why risk military humiliation when political victory remains possible? Participants found themselves naturally gravitating toward strategies that minimized operational complexity and maximized the potential to de-escalate if things went badly wrong and then re-escalate later at a more favorable time.

Read more

2. Tao of tech

In other “what’s China thinking?” news, a new piece by Afra Wang in Asterisk Magazine explores the intellectual influences shaping China’s tech elite. Apparently, it’s a bold mix of Silicon Valley’s startup canon (Jobs, Thiel, Wired’s Kevin Kelly) with Maoist tactical texts, classical Chinese philosophy, and literary works. We avoid learning more about the deeper foundations of the Chinese tech sector at our peril.

The red canon and Silicon Valley canon never contradict — they coexist seamlessly within the same enterprise. Huawei, as one of the world’s most successfully globalized companies, appears intensely patriotic precisely because of this effortless code-switching between intellectual frameworks. Internally, Huawei has nearly replicated IBM’s Integrated Product Development process wholesale, adopting thoroughly Western management practices. Yet from the outside, you would never detect the company’s profound absorption of Silicon Valley orthodoxy. This dual fluency — speaking the language of revolutionary struggle for internal mobilization while implementing McKinsey-grade operational excellence — represents a uniquely Chinese form of corporate bilingualism. The company’s public patriotism masks its private cosmopolitanism.

Read more

3. Rebooting Damascus

Over in Rest of World, Emily Wither has a fascinating interview with Syria’s new communications minister, Abdulsalam Haykal, who faces the daunting task of rebuilding the country’s devastated tech infrastructure after 14 years of civil war. Haykal’s “digital Silk Road” plan aims to transform Syria into a data corridor through private sector investment.

Submarine cables carry traffic through the Red Sea to Asia, and this route carries 17% of global internet traffic, and 90% of the Europe-Asia traffic. This is a choke point for the world. Syria, through our Mediterranean coastline, can provide another route.

So if you build cable landing stations on the Syrian coast, and then you build terrestrial routes through Jordan, Saudi Arabia, out to the international cable systems in the Arab Gulf, and then the Indian Ocean, this first provides route diversity. So from a geopolitical point of view, this is something that countries are seeking, hyperscalers [companies that operate large cloud and data centers] need, and so on. But also, the Syrian route would be shorter by around 6,000 kilometers (3,700 miles) for a round trip.

Read more

4. Model uncertainty

You should also check out David Stainforth’s latest in Aeon. David writes that the earth system models we use to predict our climate future are too flawed to make reliable forecasts. The climate science community continues to rely on them, though, investing billions in making them higher-resolution rather than addressing their fundamental limitations, which we may never overcome.

If we understand the uncertainties well, we can bring our values to bear on the risks we are willing to take. Uncertainty therefore needs to be at the core of adaptation planning while also being the lens through which we judge the value of climate policy and the energy transition. In my view, climate researchers and modellers wanting to support society should focus on understanding, characterising and quantifying uncertainty, and avoid the trap of seeking climate models that make reliable predictions. They may well never exist.

Read more

5. Hire wire

Finally, as we think about the future of AI and jobs, it is worth taking a moment to step back and consider the stories of some individuals who are already feeling displaced. “Notes From an Unemployed New Grad Watching the Job Market Break” chronicles a computer science graduate’s struggle to find work in a “broken” job market, where entry-level positions are disappearing as companies increasingly replace junior workers with AI tools, offshore teleoperation, and automation. Despite low official unemployment, the writer posits, an entire generation faces a reality where doing everything right no longer guarantees a foothold in the economy.

I do not know how many jobs will exist in twenty years, or whether my own work will sit far enough into the tail of the distribution to matter. I will certainly try to become an out of distribution human by doing a lot of different things, and by refusing to live entirely in the centre of the curve but if your entire life plan rests on being a respectable, central case worker, doing a standard job in a standard company, I think you should at least stare straight at how much effort is going into eroding that category. If your politics rest on the idea that everyone will work full time and find dignity there, you should stare at it too. The twentieth century spent a lot of intellectual and moral effort glorifying labour because economies needed people to show up every day. The twenty-first century is starting to build machines and systems that do not need quite as many of us.

Editor’s note: After the publication of our newsletter, this recommended post was taken down.

From Riskgaming

Share

From around the web

1. The seven habits of highly defeated people

Is there anything more American than the sunnily optimistic self-help book? Probably not, but even those seem to be going extinct. In The Drift, historian Erik Baker critiques the new genre replacing the old: “self-help philosophy” books that teach acceptance of failure and limitations. In some ways, books like Mark Manson’s “The Subtle Art of Not Giving a Fuck” may be a necessary corrective to toxic positivity. But they ultimately encourage political disengagement by teaching people to accept rather than challenge the structural forces causing their suffering.

While an exaggerated sense of our own importance is a recipe for both political and psychological disaster, it is also possible to overestimate our insignificance. Acceptance shades easily into excuse. With enough practice tolerating imperfection you can learn to forget what it is you’re failing to live up to. When those expectations include owning a vacation home or winning a Nobel Prize, letting go might be healthy — but doing so is tragic when they include stopping a genocide or ending homelessness.

Read more

2. Without a Voice

Whatever the consequences of our newfound Stoicism, it was perhaps impossible to sustain optimism in a culture that feels so stagnant. Celine Nguyen looks at the issue through the lens of cultural criticism. The problem isn’t that today’s artists lack talent, she writes in Asterisk, but that professional criticism has collapsed — exemplified by the demise of institutions like the Village Voice. The solution isn’t nostalgic revival of the past but recognizing that vibrant artistic cultures require a “critical mass” of people—not just professional critics—actively engaging with, critiquing, and championing new work.

What would it take for web-based art to become a significant art movement? For more people, I think, to pay attention to these works, write about them, and contextualize them in existing cultural narratives. To take part in making them, and to participate in critiquing them — pushing both artists and audiences to expect more. These are new forms; they need their critics and audiences. (“An audience with a high level of connoisseurship,” Fran Lebowitz once said, “is as important to the culture as artists.”) Here I’m reminded of a proposition from the art professor and critic Alex Kitnick, who, at a March event hosted by the arts criticism website 4Columns, quoted a work by the poet David Antin on the New York art world of the 1960s:

There were millions of people all around and most of them seem to have been artists of some sort or another … artists went to see each other’s work and we were all very excited with everything we were doing and we were all doing everything … and everybody was an art critic … all artists were art critics.

Read more

3. Controlled burn

Something more frothy than cultural creation (and perhaps one reason some artistic expression feels so flat) is AI. But most signs point to precariousness there, too. Entrepreneur offers a useful framing of what’s in store: AI isn’t experiencing a traditional bubble but rather a “wildfire” — a necessary correction that will clear out weak startups while leaving infrastructure and talent for stronger companies to absorb. His key insight is that productive wildfires destroy paper value but create durable infrastructure — and companies that can prove unit economics and genuine customer value now will dominate once capital becomes scarce.

This creates a dangerous feedback loop. Each massive spending announcement forces competitors to spend more, which drives up the perceived stakes, which justifies even larger commitments. No executive wants to be the one who underinvested in the defining technology of the era. The cost of being wrong by spending too little feels existential; the cost of being wrong by spending too much feels like someone else’s problem — a future quarter’s write-down, not today’s strategic failure.

It’s precisely this dynamic that creates productive bubbles. The rational individual decision (match your competitor’s investment) produces an irrational collective outcome (vast overcapacity). But that overcapacity is what seeds the next forest.

Read more

4. Hot takes, cooler data

Another reframing: In , Ted Nordhaus offers a corrective to today’s climate discourse in line with last week’s recommended reading from Bill Gates. It is time, he says, to revise downward our views on climate risk. Worst-case warming projections have dropped from 5°C to 3°C, and climate-related deaths have plummeted despite 1.5°C of warming. Ted argues that progressives cling to catastrophism due to social pressure, professional incentives, and the belief that only existential fear can justify rapid energy transformation, creating an insular discourse that misrepresents climate science just as badly as climate skeptics do.

Despite close to a degree and a half of warming over the last century or so, global mortality from climate and weather extremes has fallen by a factor of 25 or more on a per capita basis. As Pielke documented recently, the world is on track this year for what is almost certainly the lowest level of climate related mortality in recorded human history, not only on a per capita basis but on an absolute basis as well. The economic costs of climate extremes continue to rise, but this is almost entirely due to affluence, population growth, and the migration of global populations towards climate hazards, mainly cities that tend to be located in coastal regions and flood plains.

Read more

5. At gunpoint

On a different note, Deputy Defense Secretary Stephen Feinberg, a private equity billionaire, is driving Pentagon procurement reform with intense scrutiny of both old defense contractors and new venture-backed startups. While there’s bipartisan consensus that defense acquisition needs modernization, Feinberg’s cautious PE approach clashes with the risk-taking VC culture of Silicon Valley defense firms, creating tension as he centralizes control and demands proof that startups can deliver in actual combat environments.

Fixing problems like these was, after all, the logic of bringing in a private equity guy: He spent his career finding value inside rusty old companies like Safeway, ruthlessly reorganizing and purging them, and selling leaner and more effective versions of them back to the public markets. Cerberus was known for its discipline — squeezing out nickels where it could — and, famously, for dodging a bullet when it dropped out of the bidding for the real estate company Equity Office in 2007, just before the crash.

Now, Feinberg is applying that logic to a $300 billion-plus defense industry with, effectively, a single major customer: the US government.

Read more

6. Tour de force

Finally, what’s better than a human? A human with a bicycle (not a joke). We’re typically inefficient walkers because we must constantly fight gravity with each step. Bicycles transform us into some of the world’s most highly energy-efficient travelers, making locomotion on land more like swimming—where animals glide efficiently through water. H/t scientist-in-residence Sam Arbesman.

Read more

From Riskgaming