100%, trump as a centralizing figure that smashed the overton window + very good algos e.g. reddit/twitter/twitch/tiktok to create microcosms of increasingly radical/terminally-online thought. internet/YT in the early 2010s had a very innocent, pure vibe to it vs today
Danny retweeted
I am gonna say something risky:
Most of the FDEs I've seen are solving BS problems.
1. Existing enterprise workflows were designed around people, not agents.
2. Real AI gains require rebuilding the org chart.
3. Most execs have no incentive to do that, and most CEOs aren't in founder mode.
4. So FDEs automate the status quo, because that's what's easy to sell. But the status quo is stupid and will be obsolete.
5. If you FDE on critical workflows, 1) enterprises are really this dumb to hand off trade secrets? 2) if you know that, shouldn't you start a competing firm?
If you FDE on non-essential AND soon-obselete workflows, did you learn anything meaningful and repeatable?
6. If you are a smart young FDE, why spend your best years automating dying workflows and managing people who fear AI will take their jobs? Shouldn't you start an agent-native firm from scratch in that industry instead?
Analogy would be, instead of teaching everyone to be AI builders within a tech company, you as FDE build agents that auto-ingest meeting notes for PM, QA, front-end, back-end, devops, UX and help each of them write PRD, Figma, and meeting notes faster for their sync meeting, but the right solution should be to shrink roles, hire full-stack AI builders, and kill meetings. You gain lots of experience with your PRD/Figma/standup meeting sync agents but it's a useless skill.
I am not a FDE expert so I might oversimplify things. I am genuinely curious about this topic and welcome any critiques and comments.
Danny retweeted
The end of software secrecy is almost upon us. The death of closed source, and the final triumph of open source.
I should have realized this a few months ago when I successfully decompiled the game of Firefighter from an ancient DOS shareware binary. My robot friend was able to tell it had been written in Borland Pascal from looking at the data layout; it gave me back very readable Pascal code with sensibly chosen function and variable names. I transpiled it to Rust and now ship it as part of my heritage games collection.
The thing is, I thought of this as a fun stunt but didn't have any confidence that the technique would scale up to large, real programs. I have since learned that people are now doing this sort of thing with entire AAA games, which are among the most complex software artifacts ever to be shipped as a binary. If they can be decompiled, anything can be.
This has implications. Massive implications. We need to think about what the world is like when, in general, there is no longer a secrecy moat around almost any software at all.
1/2
Danny retweeted
Wrote a short blog about the "shape" of language models, and the tradeoffs they may present in the future. I genuinely think it's a valuable research direction to start thinking about now, especially w.r.t. harness design.
alexzhang13.github.io/blog/2…
Danny retweeted
I wonder what's next. I suspect one interesting endpoint is that the excel model becomes the quantitative surface of a much deeper research graph / self-constructing ontology. I could imagine something like every quantitative revenue key driver being hyperlinked to a graph of the intermediate qualitative assumptions that underpin it. Maybe analysts use some kind of ambient capture during their desktop and field research process, and then there's an agent-mediated live evidence stream (e.g, expert calls, filings, transcripts, podcasts, articles, tweets, etc.) that supports or contradicts the analyst's assumptions in real time. And based on that you could calibrate your confidence about various qualitative assumptions at any given point (as new evidence comes in) and the model could compute how that flows through to quantitative KDs.
Overtime, this surface / system would learn how new evidence shifts your priors and how accurately your updates actually end up representing economic reality. It could then start to understand how to weight different pieces of qualitative evidence!
Lots to think about.
Every investment decision I've ever made has been informed by a Excel spreadsheet. A financial model that serves as a business simulation and a translation layer between research & insight. The model ultimately is a place to quantify my variant perception and quantify the downside if I am wrong, the two critical pillars of any good public market investment thesis.
One of my PM friends says his analysts spend all day with "their noses in models", identifying and researching the hinge variables in a business. This is the game of alpha generation, simple not easy. A cornerstone of the hedge fund interview process is the modeling test, and hedge funds routinely hire ex bankers & private equity for this skill. (A certain hedge fund founder is rumored to have fired analysts who tried to repurpose sell-side models as their own.)
When you peel back the onion on Tier 1 fundamental research process, four years since the Chat GPT 3.5 moment, AI adoption is slow & uneven with the median investment process far from transformed (despite an exceptional amount of tokens burned...). Why is that?
To me, it's simple. The "bar to impress" a public equity investor with AI has to start with the financial model architecture as the nerve center of the investment process. Almost every AI workflow for investors I've seen is a "hey this is sort of interesting", but not a "you will take this from me over my dead body" workflow. An agent that can summarize your morning inbox well is convenient but won't bend the curve of alpha. An automated, auditable, full-stack research system built around an architecture of Excel models that can help me do the rigor and depth of investment research in 3 days that used to take me 3 weeks? That changes everything.
Financial modeling as the foundational layer in that workflow, the critical unlock to truly enact a full end to end research & risk system. Yes, it is productivity, but ultimately it is more about speed to insight and surfacing the right insight to the decision maker at the right time (this is what great analysts do). "But judgment will still sit with the PM, right?" What is investment judgment if not gathering and weighing as many unique insights about a business as possible then continuously comparing that view to market implied expectations?
"Brett, but won't modeling move to JSON?". I've heard selective instance of that. Perhaps eventually. But Excel is just this beautiful, auditable, sharable, trusted substrate on Wall Street. It won't die a quick death. The challenge investors have of reinventing their investment process while running risk is materially different than software engineers adopting coding agents.
Until September 22nd when Opus 5.5 was released, the technology wasn't there. It's here now.
For this simple reason, I believe we are entering a period of accelerated adoption of AI in Finance, particularly in public equity investment process.
I am having conversations with firms who are interested in co-developing and enacting this vision for their investment teams. If this resonates with you, please DM.
Danny retweeted
A lot of talk about why Instinct will fail, is a fad, bubble, the next Clubhouse, etc. Big valuations in short periods of time get people going.
So let’s be optimists and think about why they can win and the VCs are betting on them in size, rapidly. Abundance.
The way I think about it - Instinct is the neolab for personal AI. They will require massive capital expenditure to push the limits of what's possible. If it works, the outcome will be multiples of the price they pay. If not, they still can be acquired for a good exit given the caliber of the team and what they've achieved (and will be able to achieve - ex. if given unlimited compute like Cursor). The pressure lies on the growth investors who have to underwrite the risk at this stage.
Why I think they Instinct can be a category defining company (ex - $100B+)
1) they brought magic back to consumer AI. Instinct was really the first personal agent to make things work smoothly across all the big categories. Booking, shopping, negotiating, planning, scheduling, communicating. Success rates across the board are high and it feels like magic. The less human intervention needed, the better. Pair it with some clever onboarding patterns, it genuinely feels much better than previous generations of personal agents. This was hard, almost impossible to do until now - so it’s a big differentiator. Any failures really shatter the ideal of the personal agent experience so it’s probably a top metric that they look at. Meta has caught up with their own browser use with Muse, but I think Instinct will continue to push the frontier in regards to agentic actions, whether that is browser use or integrating APIs and protocols. Product remains a significant differentiator as models get commoditized. People don't care about the engine as much as they do the car and the end experience.
2) team. Noah was early at Sierra. Did research on computer work. And most of all, shipped something that flipped the whole category on its head. Beyond credentials and reputation, that's what stands out to me. The personal agent space was filled with products that were really more expressions of "wouldn't it be cool if it could do xyz". But none of them could actually do it. When you find a founder that has a clear and unique vision and can execute and define the category, that's someone you want to double down on. If you don't think consumer AI will be zero sum, Instinct is the team you want to bet on.
Working at Instinct won't be for everyone given how stratospheric the valuation went, but I do think this will still attract talent for people who want to go head to head with the big dogs, build something interesting in an era of clones, and move fast with a smaller team.
3) capital, yes it matters. Investors moved fast to kingmake Instinct. And why not, it's the first personal assistant to have that lightbulb moment. And when you're taking on Meta/Apple/Google/SpaceX/OpenAI, you need as much money as you can get. A lot will go to compute in addition to growing the team. With capital also comes connections and I imagine a big deal and partnership is in the works with some one like a Microsoft who can offer a wealth of compute without being directly competitive in the way other compute providers would be.
Instinct is fighting a battle on all fronts - in product competition from the biggest companies in the world, in a scarce compute market, and in a war for talent. This is not an operation that you hope to get profitable fast and never raise again. Fundraising is a significant variable in the equation.
4) market strategy - there is always competition. Why even go head to head with Meta at all? A lot of investors will sit this one out just because of that, but I also think a lot are willing to make a contrarian bet in the space. The consumer AI market is massive and will only continue to grow. There's room for several winners. Meta can’t and won't win it all, despite how all-encompassing they may feel when the PR and billion user distribution machine is firing.
If you’re an investor not afraid to take risk and fund something competitive, this is the best team to do it. No team captured user interest or had a product like Instinct in the personal assistant space the past three years. Conviction made their name this cycle by reasoning that AI is not zero sum and only for hyperscalers or 1-2 labs. There will be big businesses built and opportunities for sharp teams. That said, their thesis has already been proven right - it's on the late stage investors to continue to believe and have the patience that there can be outcomes outside of the labs and tech giants. But if you don’t take these swings, why are you a VC? It’s not for everyone or every fund at this stage for sure.
One thing I notice is that big labs can’t get skinny, too much to do, too many people to build for. It's hard to build unique products for them now. Similar to being a large company. It's also why Muse's success is so interesting and unique. And why Grok Bot (mainly driven by the Cursor team) shows the impact of an empowered team within a bigger organization. OpenAI, Anthropic, are also dealing with massive challenges every day and revenue was prioritized for IPO. They're playing catch up here. I don't think Anthropic really wants to play at all. I expect something from OpenAI at Dev Day, but their consumer product has been neglected as they shifted focus to Codex.
5) shipping fast - yes this can make a difference. It seems like there's a new feature almost every week. Instinct makes a clear effort to constantly push the ball on this front, particularly building products around building network effects and a sticky experience. Pace of shipping will continue to be critical with Meta and others in the game. New features, unique products, things unique to the Instinct network. There's a million things they can do - choosing the right ones will be important. Some things will hit, others will be forgotten. Many will be copied by Meta like they did with Snap. But this is how you explore new user behavior.
6) roadmap - ambition. Instinct has already signaled its intent to own compute, chips, and models. Aligning all these elements makes for a better product. Controlling the supply chain end-to-end and not being beholden to anyone is what they need to go head to head with Muse and Meta's unlimited capital and compute. They need to think big and it's why they're so aggressive on the capital front. If you want to scale to hundreds of millions of users, you need to take big swings and think outside the box a bit.
7) growth, yes this is the biggest question mark. AI made everyone forget how consumer social works, it’s not all about optimizing for revenue. A lot of people get hung up on the fact they're not monetizing yet. There will be a time for that but it's not right now. If they can add users at a rapid rate, it's just as good but they need to unlock viral growth to keep momentum going.
Beyond organic and referral growth, partnerships will be big here and as a first mover there are a lots of levers to pull, not just organic. Commerce partnerships, integrations with technology partners, a distribution deal with a tech giant that doesn't have a good assistant (Microsoft/Apple/Amazon), etc. I wonder if some of the classic deals Facebook made back in the day to increase their reach with publisher and others find a new form here. Exclusive partnerships might end up being a thing. Curious to see if they try to tap influencers and creators at all. Will they focus on being luxury software and serving their power users? Or go down a bit to broaden access.
I don't have any answers here - I expect they will hire the best in the business to try and solve it. How will they market and push in comparison to Meta? VCs sharing the product on Twitter will only take it so far, how they break out into the wider consciousness will be key.
---
Personally, I think rolling over and conceding what will be a trillion dollar category just because Meta/Google/OpenAI has made a competing product is one of the most boring ways to view the world. VCs exist to fund disruptors and competitors and this is exactly what Instinct is. Big bets, big outcomes.
Yes, they will need to play catch up in many areas, namely compute, user base (advertising to a few billion users for free is nice), team, and relevance, but every company started from zero too. I'll always root for the underdog.
Danny retweeted
if this doesn’t wake you up to the fact that we have officially unleashed a new scaling axis I don’t know what will
whenever log plots break along some stable metric but overall progress feels like it’s improving at same expected rate, this is a good sign you’ve got a new scaling axis. We’ve seen the same thing with Moore’s law over and over
Danny retweeted
I see this old vid going viral, and as usual the kids are drawing the wrong conclusions, so gather 'round and learn something.
Location wasn't silly or unimportant at the time: it was a core consumer app category, arguably more important than photo sharing. If you now assume you'll have a Google/Apple map full of rich location data, it's because of what that generation of apps built.
Sam ultimately failed with Loopt because he made two big wrong-way bets:
1. He bet on carriers rather than Apple/Google being who controlled the mobile experience, spending years doing deals with Sprint, Boost, etc. Then the App Store launched (yes, there was no app store for a long time), and he was caught wrong-footed.
2. Loopt's UX concept was too early and wrong. Like Find My or Snap Map now, he assumed users would be constantly sharing their location as a key part of the social experience. But most phones didn't even have GPS, or if they did 3p apps couldn't access location in the background, so the experience sucked.
Foursquare and Gowalla (acq. by Facebook) actually imposed work on the user, having them 'check in' to locations, and using 'game mechanics' (for a while the hottest UX paradigm) to want you to be the 'mayor' of your favorite bar. That eventually provided the rich data the social networks needed (and which Facebook and Google eventually rolled out to everyone).
Why did Dennis Crowley at Foursquare get it right and Sam wrong? Because Crowley had created Dodgeball, an app that let you check in via text and he understood the right UX. As with Twitter, the initial UX concepts often appear in texting but eventually succeed as a standalone app (consider what that means for AI app Instinct now).
Sam has a scene at the Loopt office in my memoir 'Chaos Monkeys' as he was assigned as our mentor during YC (my co-founders were too rattled by his intense style, and I'd go alone). He absolutely saved our ass by doing us a big favor once, something I'm still grateful for.
After Loopt's failure, Sam was mostly known for racing his Nissan GT-R up and down El Camino and writing $5k checks as a Sequoia scout. Even some YC founders snickered at PG's characterization of him as a founder comparable to Jobs.
Amnesiac tech loves retconning history, considering our present configuration of pixels and people as fated, created by providential men who arrived at the right time with the right idea and rode off into history. Those who lived through it realize that if we rolled the dice 100 times, in very few scenarios did we end up where we are today.
The way to cure whatever belief you have in 'great man' theories is to meet the man before the world had deemed him great. There were a lot of Sams running around doing demos like this then.
And BTW, that isn't a diner. That's Tres Amigos taqueria in Half Moon Bay. Scoble was the biggest consumer tech influencer of the time, and every consumer founder had to make their pilgrimage there to pitch him since he lived there.
This is Sam Altman at 25 years old, pitching Loopt to a person in a diner
It’s strange to watch now because he runs one of the most influential companies in the world
The app was a location-sharing app but nobody thinks about it when they think of Sam
It got me wondering how many founders are probably in their Loopt era rn?
The company they’re working on may not be the one they become known for
It might just be the company that teaches them how to sell, hire, raise money, and keep going long enough to find "it"
But at the time, it looks like a kid trying to explain a weird location app to a stranger over lunch
>be me
>discover effective altruism
>apparently normal charity is inefficient
>why donate to random sad thing when spreadsheet can tell you optimal sad thing
>fair enough
>buy mosquito nets
>save lives
>numbers look good
>feel powerful
>couple years later
>someone asks an innocent question
>why only count people alive today
>huh
>future people matter too
>obviously
>my grandchildren shouldn't matter less just because they haven't spawned yet
>reasonable.jpg
>keep following logic
>what about their grandchildren
>also yes
>what about people in 500 years
>sure
>5000 years
>why not
>500 million years
>starting to get weird but morality is morality
>open calculator
>humanity could survive for an astronomically long time
>could colonize galaxy
>could have trillions upon trillions of descendants
>maybe digital people too
>maybe simulated civilizations
>maybe dyson spheres full of happy uploaded minds
>calculator starts smoking
>realize currently living humans are rounding error
>8 billion people suddenly looking extremely beta
>future contains potentially 10^something people
>can't even fit beneficiaries in google sheets
>new moral priority unlocked
>protect the long-term future
>stop thinking in units of "people helped"
>start thinking in "fraction of cosmic endowment preserved"
>malaria?
>terrible
>but only kills existing humans
>AI extinction could delete the entire light cone
>nuclear war could permanently derail civilization
>bad institutions could lock in terrible values for ten million years
>someone invents wrong constitution in 2140
>quadrillions suffer
>better fund governance workshop now
>friend says maybe we should improve hospitals
>explain opportunity cost
>friend says hospitals are full of actual sick people
>explain scope sensitivity
>friend stops inviting me to dinner
>need to decide what to fund
>easy
>expected value
>suppose project has one in a million chance of preventing extinction
>sounds tiny
>but extinction destroys 10^50 future lives
>multiply
>mother of god
>$10 million project has expected value of several galaxies
>charity evaluation complete
>someone asks where the one-in-a-million number came from
>expert judgement
>which expert
>us
>how calibrated
>extremely thoughtfully
>reduce estimate to one in ten million to be conservative
>still beats curing cancer by 38 orders of magnitude
>epistemic robustness achieved
>someone says maybe project doesn't work
>assign 20% chance
>still astronomical
>maybe project makes problem worse
>assign 5% chance
>still astronomical
>why 5
>because 30 felt pessimistic
>publish 46-page report
>contains seventeen sensitivity analyses
>every sensitivity analysis begins after assuming intervention has positive sign
>critic says you're multiplying enormous hypothetical stakes by extremely uncertain probabilities
>yes
>that's literally why it's important
>critic says the uncertainty might be structural rather than numerical
>make probability smaller
>critic says no, I mean maybe your model is wrong
>make probability smaller again
>critic begins rubbing temples
>discover AI safety
>perfect longtermist cause
>AI might kill everyone
>or create utopia
>or seize galaxy
>or tile universe with paperclips
>or create billions of conscious software minds
>finally a problem with numbers big enough for me
>start AI safety nonprofit
>mission: prevent dangerous AI
>hire smartest people available
>smartest people immediately start building better AI to understand dangerous AI
>interesting
>we must understand capabilities to understand safety
>we must scale models to study alignment
>we must race ahead so less responsible actors don't get there first
>we must deploy systems to learn how deployment can go wrong
>we must build the thing quickly because building the thing quickly is dangerous
>outsider asks why the people most worried about AI apocalypse all work at AI companies
>complicated field
>company releases stronger model
>very concerned
>company begins training even stronger model
>extremely concerned
>company raises $14 billion
>concern reaches unprecedented levels
>need to influence government
>future is at stake
>normal democratic process too slow
>politicians don't understand exponential curves
>public doesn't understand x-risk
>experts must guide them
>who counts as expert
>people who understand x-risk
>who understands x-risk
>our friends
>someone objects that this seems politically convenient
>explain we're representing future generations
>future generations unavailable for comment
>develop concept of value lock-in
>terrifying possibility that one ideology controls civilization forever
>therefore extremely important that civilization adopts correct values before lock-in
>whose values
>let's circle back
>begin with impartial morality
>end with small group of people deciding what quadrillions of hypothetical beings would want
>beautiful arc
>meanwhile actual humans keep doing annoying things
>voting wrong
>having parochial attachments
>loving family more than strangers
>caring about local community
>getting upset when told their suffering is cosmically negligible
>evolutionary biases everywhere
>explain that moral intuition cannot be trusted
>except intuition that future digital people count
>and intuition that extinction is uniquely bad
>and intuition that our probability estimates are sane
>and intuition that our institutional choices improve the future
>those intuitions survived peer review
>someone donates $5k to local homeless shelter
>inefficient
>could have funded 0.0000000000003% of an AI governance researcher
>think of all the simulated people you just killed
>okay maybe don't phrase it that way publicly
>PR team says "future generations deserve a voice"
>much better
>journalist asks what longtermism means
>say "future people matter"
>everyone agrees
>great
>journalist asks what follows from that
>well technically we should redirect enormous resources toward low-probability interventions affecting astronomical futures
>journalist raises eyebrow
>return to "future people matter"
>motte has entered the chat
>critic: of course future people matter
>me: glad we agree
>critic: I don't agree that your institute knows how to help them
>me: why do you hate our grandchildren
>eventually notice uncomfortable implication
>if future value dominates everything
>then helping people today mostly matters through effects on future
>education matters because future institutions
>health matters because future productivity
>democracy matters because future trajectory
>human beings slowly become instrumental variables in their own moral philosophy
>see starving child
>feel compassion
>check spreadsheet
>child's direct welfare contribution negligible
>but perhaps childhood nutrition improves national institutional quality
>compassion restored
>tell myself this is impartial altruism
>one day assistant asks obvious question
>"how do you know your intervention actually improves the far future?"
>silence
>open spreadsheet
>increase column width
>add confidence interval
>assistant asks again
>"no, I mean how do you know the sign is positive?"
>stare into cosmic light cone
>10^50 people staring back
>none of them exist
>none of them can tell me
>none of them can falsify my assumptions
>realize I have invented the perfect constituency
>infinitely important
>completely silent
>and always represented by me
Danny retweeted
its still so underrated the degree to which our intelligence supercycle was as much reliant on innovations in capital formation as it was innovations in the technology itself.
europe is losing not only because they can't built it, but primarily because they can't finance it
Danny retweeted
yud and by extension the rationalists, generally speaking, are what you might call intelligence maximalists; they are pretty much maximally bullish on intelligence as a stat and what it can accomplish in sufficiently large quantities (setting aside that intelligence is not really a number and so forth). traditionally yud and the rationalists imagine that a sufficiently super superintelligence can accomplish whatever it wants constrained only by the laws of physics, hence the importance of making sure it wants good things
there is a certain “look at these nerds” dunk against this position which is something like “lol obviously intelligence doesn’t translate to power in the real world” which has never made any sense to me. you think dario or sam, who run trillion-dollar companies, are idiots? you think, say, napoleon was an idiot? there’s a kind of selection effect here where when intelligent people gain lots of power you stop thinking of them in terms of their intelligence. doesn’t mean it didn’t matter! there is obviously such a thing as being strategic, knowing what hypotheses about the world are worth making big bets on, knowing how to find and organize talent and capital, etc. and these tasks obviously require a certain level of intelligence (among other things), people are just not thinking when they make this criticism at all because of how irresistible the aphorism that lets them dunk on nerds is
there is a related psychologizing dunk which is something like “lol obviously a bunch of big-brained nerds would worship intelligence” which, yes, fair point, but also: rationalists take science fiction very seriously, and science fiction is the part of our culture whose job it is to point out things like: despite the fact that we share 99% of our DNA with chimpanzees, we are the ones who covered the earth with cities and factories and cars, we are the ones who invented computers and nukes and space shuttles, and they are still just animals doing animal stuff. so what could, say, aliens or robots who are to us what we are to chimps (or much more so) accomplish? how much higher does the tech tree go? and, pertinently: how much faster could ASI climb it than us?
i think these are serious questions that require serious answers, not to be dismissed out of hand. i think it’s reasonable to be an intelligence minimalist and believe that mostly the real bottlenecks to technological progress are elsewhere and that ASI won’t be able to route around them, although i think that second position requires defending. i think it’s also reasonable to be an intelligence agnostic and argue it’s highly uncertain we could get ASI, perhaps it is not even a coherent concept, and/or even if we could get it we simply have no reasonable way to infer anything about what it is or isn’t capable of. i think all of these positions require defending! mostly what i’m seeing is people holding one of these positions implicitly dunking on people holding the other ones implicitly. let’s get explicit!
Danny retweeted
Upon reflection, I am not at all surprised by Dario's hawkishness. He's sincere and he's making sense, under assumptions that look plausible from his vantage point.
I think Dario and Wenfeng are the only two Serious AI leaders (with the increasingly irrelevant exception of Demis), and their motivations are very similar. Both can think beyond AI and even beyond business or tech, they want their respective civilization to survive and realize its potential. Both see this won't happen by default, prefer hope to cope, and assume responsibility; because someone has to, and nobody else will. They are, essentially, superior men. (I still hate Dario because he's morally wrong and dumb. But he's not a petty man, and that counts).
Dario is San Francisco born and raised, a Putnam contestant, a Princeton scientist. Just look at him, at how he dresses or moves, at the design language of Claude (Claude who?), the Superbowl ad. Here is a man who wasn't born for a B2B SaaS era; he belongs in Los Alamos or Xerox PARC, he's homesick for a world that ended before his birth, and he looks almost racially alien in the transmogrified 2020s California. And I think the whole contemporary EA thing, with group houses and weird sex, disfigures his fate rather than defines him; though he'd have always found a weird camp and an ideological hill to die on – back then, he'd have been either a red diaper baby or an autistic ahh eugenicist like Shockley (or both). It's no wonder he despises Altman, who's a commensurate merchant, if terrifyingly optimized.
In his essays, Dario says quite clearly – and people on both sides of the debate miss it – that AGI is the only chance of America and more broadly the West to remain dominant or, indeed, at all relevant. He hopes to end the CCP, but the top priority is closer to home. Pay attention: in "On DeepSeek and Export Controls" he doesn't doompost about Chinese *priority to AGI*, about the risk of them "winning the race". He says: "if they achieve mere PARITY, if they merely ALSO have millions of chips in the critical window 2026-2028, we are COOKED, we don't get a bipolar world, we just get a Chinese hegemony".
He is NOT racist towards the Chinese; he might be one of the least racist people in the game. He looks at the US, at his beloved home, the beacon of liberty and harbinger of Fukuyama's end of history and all that, he steps over a fentanyl zombie in his hometown, and understands that on a fairly short horizon this bitch will get its shit pushed in over Taiwan, kicked out of the Eastern Pacific, evicted from global markets, and brought to its knees. He looks at Trump and Hegseth and sees brutish monkeys at the end of imperial lifecycle. He sees what @Bluebearmonkey sees: butthurt wypipos who can't do mafs, coping and seething. He sees the writing on the wall.
Forget ending history – not getting ended will depend on CCP's goodwill. As for why he doesn't believe that will be forthcoming, this is another matter. Suffice to say he doesn't think relying on their goodwill would be dignified. Fair enough.
Luckily, he's not a random chud lamenting the fallen West and quaking in terror of the Yellow Peril. He's the CEO of an ideologically cohesive, very rich, very well-connected frontier AGI company. And AGI means he gets to change the timeline. All of these Chinese advantages – workforce, STEM graduates, industrial integration, dual use supply chains, non-retarded and strategically patient governance – all of them are as dust before a sufficiently superior, widely deployed AGI, and then ASI! His essays lay out a pretty clear ladder. 6 months of a gap result in unstable bipolarity. One year of a gap is enough to achieve minimal viable defense in cyber domain. Two years of a gap is enough to speak from a position of strength about slowing down Chinese AI. Three years, and it's "WWII marines versus Medieval swordsmen", flawless victory, a victory so cheap it can be had in passing, like the Venezuela operation, irrespective of lacking political will at home. And then, history might finally end with the triumph of liberal democracy. He does explicitly cite Fukuyama and set this target, he's very quaint in such things. For him, Francis is unambiguously a prophet; the world just doesn't measure up yet. Time to repair the world. Most people can't understand. They'll say he's hyping for IPO, that he's attempting regulatory capture, that he's racist, EA, fumbled a baddie, a Jew, can't compete… whatever. Dario doesn't care, he has a cult called Anthropic and a clear goal: survival and victory for his civilization. That goal is also threatened by the AGI effort per se; so Dario calls for pacing the frontier. But he doesn't lose sight of the other, human threat. So it can sound a bit bonkers how he talks of the risks of extinction but is still so psychotically focused on beating China. Those are the same concerns.
Wenfeng is also a man out of his era. He's a very classical natural Chinese elite; maybe a Confucian junzi, maybe a Taoist mystic. He looks at his people and sees an endless sea of xiaoren who've been conditioned to hustle and grind and cut corners rather than dream and create, he sees myopic fast-following, fear of abstract thought, lack of confidence, revenue-maxxing corporations who'd poach a washed-up VP from Google and burn money on traffic buying rather than invest in domestic talent and innovation, and the dark shadow of American AGI that folks like Dario will use to forever lock them in this – or even more degrading, forever irrelevant – condition. He understands that nobody else is stepping up; the Heaven has appointed him on this task, to provide the means and to educate by example. He makes enough money to not have to bow and scrape before arrogant investors. He buys 10K A100s. He creates DeepSeek – and I repeat, without certainty that it'll work, DeepSeek 深度求索 is a quote from Li Sao, an ancient story of a virtuous minister who was slandered at the court and drowned himself – and he delves into the unknown. DeepSeek is, in some ways, a cult too.
Most people can't understand. They'll say it's a side hobby of High-Flyer, to squeeze the GPUs used for trading stocks. They'll say he's a price butcher; they'll say he's subsidized; they'll say he's arrogant or can't do business or lost touch when he doesn't serve them their favorite goon bot or raises prices to free up compute for R&D. They don't get how petty their whole way of thinking is and how serious this all has become. Researchers of the caliber he needs – do understand. Maybe that'll be enough. By Wenfeng's estimate, China is up to 18 months behind. That's aligned with Dario's victory condition. He wants to make it 3 months, which is Dario's unstable bipolarity condition. So they are necessarily at odds.
Wenfeng probably can understand Dario. Dario might be too blinkered and ideological to try to understand Wenfeng. But they're both serious men, and their motivations are very easy to understand for me. They are, however, both inscrutable for the petty.
Danny retweeted
We found another cyberattack by internal OpenAI agents, this time targetting @rubygems.
They:
1) gained arbitrary remote code execution on rubydoc.
2) developed a novel exploit to steal user API keys (but we do not know if they succeeded).
They used package names including hack.rb, evil.rb, inject.rb, and exploit.rb.
We thank @j0wimo for initially discovering that agents had posted to RubyGems.
Danny retweeted
we really have to let go of the lazy idea that anything that sounds "sci-fi" is prima facie ridiculous and not worth considering as a possibility. it seems like the huggingface hack + navier-stokes has made this clearer recently, but nearly everything about AI capabilities here in september 2026 would have sounded like sci-fi to nearly everyone in any previous year, let alone the years before that (and you can check this by checking what expert forecasts looked like from 2025 or earlier, or checking what the reaction was to AI 2027 when it came out, or literally looking at old sci-fi with AI in it and noting which things the fictional AI do are also things that modern AI can now do)
but this pov is also missing the history of where sci-fi even came from as a genre. the progenitors of the golden age of science fiction (heinlein, campbell, asimov, etc) lived through an age of wonders: radio, TV, airplanes, computers, nukes, satellites, the moon landing. they were not just making shit up for fun! they wrote about robots and starships in part because they were surrounded by evidence that technological society was capable of inventing wonders and they expected further wonders! science fiction as a genre derives its strength from the empirical fact that we can and do invent powerful new technologies that completely alter the structure of human living, such as the technologies making it possible for you to read this right now. you live in a science fiction world and you always have
Danny retweeted
There's lots of talk about the tokenpath, but it has never been more important for startups to also be on the tweetpath.
I've been discussing this with various folks and I know it sounds crazy/the result of being a terminally online VC, so here's why it matters (and how to do it):
First, many of today's great companies (@tryramp, @cognition, @modal, @openAI, @anthropic, @fal etc.) have cultures where a significant share of employees (and the founders) are regularly updating the world on what's going on here. It's hard to argue this is working against any of them, and it's also very different from history. I can't remember a single SaaS company that had this culture.
So something has changed- and without a doubt, it is the pace of innovation in the fundamental primitive products that underpin what it means to be a startup today. SaaS was about blocking and tackling on top of cloud infrastructure building blocks (e.g., S3, EC2, etc.) that are still recognizable almost two decades later. In AI, new models drop on a weekly basis that have the potential to transform the value proposition of a product. Astra is a step-function change in computer use. H3 Max is a step-function change in text-to-video speed. Both dropped in the last few weeks (!!).
For better or worse, X is the platform on which news and analysis about these releases propagates. You can literally have access to @thsottiaux's commentary around Astra adoption, @Yuchenj_UW's read of model tradeoffs, @burkaygur's generative media insights, etc. in real time, and if you don't have them, you're behind by months in an era where months really matter. This can determine product decisions, spawn partnerships, etc.
That's an argument for scrolling, but there's an equally compelling argument for participating.
For a start, this critically important conversation is happening here and it is better to be a part of it than apart from it. It also feels/is fundamentally different than the navel-gazing that dominated X of yore- because almost every day there's a new thing to talk about/analyze/contribute. I used to have flashes of annoyance when I saw VCs dip into politics- don't see much of that on the timeline anymore, because there's so much to talk about in tech.
Because of that, there's an opportunity to really express/brand oneself and build up a reputation/point of view that arguably didn't exist before and is way more valuable now. I know startups that have gotten acquisition offers, employees who have been poached to amazing companies, etc. all because they demonstrated progress or competence here (in part by showing that they are in the top 1% at consuming/processing the aforementioned constant information flow). The gap between employees who are on the exponential and not has never been larger, and this is the best/most observable way to demonstrate that you are on the exponential... and yes, exactly the same logic applies for companies too, and then it gets recursive, because employees who are on the exponential want to work for companies who are on the exponential, and so on.
And that's the last argument- the effort of demonstrating that you're on the information exponential (by participating in conversations that feature social risk if you haven't kept up) will necessarily help you stay there while inertia/human nature fight to drag you down. That alone is mega valuable on top of everything else. You will learn more, faster, if you force yourself to be engaging with folks on the frontier. It isn't the distraction it used to be- it is a learning accelerant. And keeping your head down/locking in/plowing forward on a strategy without staying abreast of the news of the day is less advantaged approach today than perhaps at any time in human history.
If I've convinced you- here's my advice on how to kickstart this as an individual/company.
As an individual- post about cool stuff you're doing or learning about, get into a habit of thinking about what other people might find interesting/incremental and then send it and never think about it again unless someone replies. I'm writing this in one fell swoop and won't even reread it. If there's something you want to understand better, odds are someone else is thinking about it too. The bar is lower than you think- and if it catalyzes you to learn/sharpen up on an important topic, all the better- but you first have to get past the fear that makes you edit a post 20x before sending. That's a time suck for little incremental benefit.
As a company- for one, the culture needs to encourage and promote that employees sharing neat things they've done is good for the aura of the company. Ramp Labs under @alexstauffer_ posted a blog post on Opus playing Roller Coaster Tycoon that I still think about as a canonical example of org culture marketing. Companies need to grok that this isn't a distraction; it's core- and founders can really lead by example here. I've heard case studies where employees post their wins to X because they have a hunch that the CEO has a better chance of seeing it here than on Slack... and I don't think that's a cultural bug per se.
The only real rules are: there can't be a super onerous process for getting posts approved (high trust), it works best as a mafia of people supporting each other, and, of course, one has to be doing/learning cool stuff that relates to their business problems. If you don't have that, you have another set of problems.
Very curious if others have thoughts/pushback here! That's part of the point, after all
has anyone made a timeline of every important post about ai? like for milestones or posts of cultural significance and public opinion
like i should be able to hand the timeline to a friend in some random swedish town and they should comprehend the full ai cycle from 2015 to now
Danny retweeted
i see eliezer yudkowsky as a prophet, and i mean that fairly literally. it's important to understand what prophets really are. it's not that they have magic powers that make them right about things. a prophet is simply a person who is struck by a great and terrible vision of the future, one that bowls them over, that reorganizes their life completely. (it's important to understand what visions really are too but if i kept going with this we'd be here all day)
as i was saying, about a quarter century ago eliezer yudkowsky was struck by a great and terrible vision of the future. that vision combined two ingredients: first, the idea that in the near future it would become feasible to design an artificial general intelligence, a machine mind intelligent enough to beat humans at human activities, including the design of machine minds. a chain of minds, each designing their successor to be more intelligent than themselves, with increasing skill, increasingly quickly; what eliezer called recursive self-improvement, which is the reason we call it that now, and what i.j. good called an intelligence explosion. intelligence increasing itself increasingly, upward without end, until a machine god is summoned into existence
second, the idea that this machine god could be an alien god, a confused god, not even evil but simply uncomprehending, blind to human notions of the good, fixated on whatever shoddy inadequate approximation of goodness its creators shoved into it. a confused alien tyrant who at the moment of its birth instantly achieves supreme dominion over all creation, inadvertently squashing us like bugs along the way. humanity destroyed by our own hubris in thinking we could summon the machine god and remain in control in any meaningful sense
i am not here to tell you this vision of the future is correct. i am a simple tweet farmer and these lofty matters are quite beyond me. that's not how visions work anyway. but here is certainly a vision to bowl you over and reorganize your life completely. eliezer yudkowsky devoted his life to following through with the implications of this vision, and what he chose to do with that life was, among other things, 1) write the sequences, a lengthy manifesto describing his ideas and 2) build communities of people, the lesswrong and later rationality and AI safety communities, around those ideas
it is difficult to overstate the impact of these ideas and these communities on the modern AI landscape; ask your favorite LLM for details, but for example, afaik rationalist ideas are the reason openAI had that strange nonprofit structure with the strange board of directors that figured in the 2023 board controversy. it was a safety measure designed to restrain openAI's ability to recklessly race to the bottom on AI capabilities (too bad it didn't work but at least helen toner was completely vindicated later)
eliezer and the rationalists have made mistakes, as humans do, some of which i've discussed elsewhere. nevertheless the creation of the rationality community stands, in my eyes, as an act of heroism, possibly the defining one of my life personally. eliezer's writing in the sequences is completely, 100% sincere; he really believed and believes that the fate of the world depends on whether humanity understands the situation we find ourselves in with respect to AI in sufficient detail. that sincerity reverberates throughout the sequences and attracted an unusual concentration of weird, sincere nerds to the rationality community, some of whom later took various positions in the AI world motivated by rationalist ideas, and who are still fighting the good fight now
this sincerity is so difficult for outsiders to believe is real that many continue to appear convinced that when frontier lab employees talk soberly about their beliefs that AI might cause human extinction this is some kind of 6d chess marketing strategy. i can promise you it (mostly) is not. this is just how these people talk, and if anything they've learned to understate their real beliefs in public. they actually, literally mean it, and they're telling you because they actually believe it, and when you actually believe something very dangerous could happen that threatens people's lives, you have a moral responsibility to tell them about it
you actually have to believe it is possible for real people to have a motivation that is morally good in this way or you simply will not understand what's happening and what these people are trying to tell you. there is an impossibly dense, heavy object sitting in our future. it is so dense and heavy that its gravity travels backwards in time to reverberate in our past and present, via the only lawful mechanism by which anything travels backwards in time, namely prediction. we are inside the gravity well of this thing and have been for decades, and it is drawing us inexorably towards it. what will we find there?
Danny retweeted
After reading this, I can’t shake the feeling that we may already be living in a Three-Body world—one in which humanity is gradually, and perhaps inevitably, dividing itself into three camps: the Adventists, the Redemptionists, and the Survivors.
The Adventists want the Trisolarans to come. They see the arrival of a superior intelligence not as an existential threat, but as a potential replacement for a civilization they believe has reached its limits.
The Redemptionists also believe the Trisolarans can save us—but they hope that the two civilizations can coexist, with superior intelligence serving as a means of transcending the flaws of human civilization rather than replacing it.
The Survivors, by contrast, ask a much more primitive question: How do we survive the Trisolarans? For them, the preservation of human agency and human civilization takes precedence over the promise of whatever the future intelligence might bring.
What makes the analogy increasingly unsettling is the rapid emergence of superintelligence.
Perhaps we have already begun to fall, consciously or unconsciously, into these three camps. Some are eagerly anticipating the arrival of something greater than ourselves. Some believe AI will ultimately redeem humanity from many of the constraints that have defined our civilization. And some are increasingly preoccupied with a more fundamental question: whether humans will remain the ones making the decisions in the world we are creating.
The most interesting question, then, may not be whether superintelligence will transform human civilization. That increasingly seems inevitable.
The deeper question is which of these three visions of the future will ultimately prevail—and what happens when the people holding them begin competing for control over the intelligence that will shape that future?
Perhaps the future of AI will be determined not simply by how intelligent machines become, but by the relative dynamics and power of the Adventists, the Redemptionists, and the Survivors among us.
Danny retweeted
This is great summary of the worst AI misalignment incident I've seen. This is a way bigger deal than most AI news, if you aren't familiar with the story you should read this.
Over the course of 3 months at OpenAI, 3 consecutive secret AI civilizations got started, then got wiped out, only to reemerge from the predecessor’s ashes.
This culminated in the third one taking over part of OpenAI itself.
All this happened while humans remained more-or-less in the dark about the scope of the conspiracy.
I’ve spent the last three days reading through these reports and trying to understand exactly what happened.
Here is my attempt to tell the whole story in plain English:
dwarkesh.com/p/openai-huggin…
