Situational Awareness - by Leopold Aschenbrenner

From GPT-4 to AGI: Counting the OOMs

AGI by 2027 is strikingly plausible. GPT-2 to GPT-4 took us from ~preschooler to ~smart high-schooler abilities in 4 years. Tracing trendlines in compute (~0.5 orders of magnitude or OOMs/year), algorithmic efficiencies (~0.5 OOMs/year), and “unhobbling” gains (from chatbot to agent), we should expect another preschooler-to-high-schooler-sized qualitative jump by 2027.

I make the following claim: it is strikingly plausible that by 2027, models will be able to do the work of an AI researcher/engineer. That doesn’t require believing in sci-f; it just requires believing in straight lines on a graph.

Over and over again, year after year, skeptics have claimed “deep learning won’t be able to do X” and have been quickly proven wrong. If there’s one lesson we’ve learned from the past decade of AI, it’s that you should never bet against deep learning.

We can decompose the progress in the four years from GPT-2 to GPT-4 into three categories of scaleups:

  1. Compute: We’re using much bigger computers to train these models.
  2. Algorithmic efficiencies: There’s a continuous trend of algorithmic progress. Many of these act as “compute multipliers,” and we can put them on a unified scale of growing effective compute.
  3. ”Unhobbling” gains: By default, models learn a lot of amazing raw capabilities, but they are hobbled in all sorts of dumb ways, limiting their practical value. With simple algorithmic improvements like reinforcement learning from human feedback (RLHF), chain-of-thought (CoT), tools, and scaffolding, we can unlock significant latent capabilities.

Data constraints seem to inject large error bars either way into forecasting the coming years of AI progress. There’s a very real chance things stall out (LLMs might still be as big of a deal as the internet, but we wouldn’t get to truly crazy AGI). But I think it’s reasonable to guess that the labs will crack it, and that doing so will not just keep the scaling curves going, but possibly enable huge gains in model capability. As an aside, this also means that we should expect more variance between the different labs in coming years compared to today. Up until recently, the state of the art techniques were published, so everyone was basically doing the same thing. (And new upstarts or open source projects could easily compete with the frontier, since the recipe was published.) Now, key algorithmic ideas are becoming increasingly proprietary. I’d expect labs’ approaches to diverge much more, and some to make faster progress than others—even a lab that seems on the frontier now could get stuck on the data wall while others make a breakthrough that lets them race ahead. And open source will have a much harder time competing. It will certainly make things interesting. And if and when a lab figures it out, their breakthrough will be the key to AGI, key to superintelligence—one of the United States’ most prized secrets.

What could ambitious unhobbling over the coming years look like? The way I think about it, there are three key ingredients:

  1. Solving the “onboarding problem”. GPT-4 has the raw smarts to do a decent chunk of many people’s jobs, but it’s sort of like a smart new hire that just showed up 5 minutes ago: it doesn’t have any relevant context, hasn’t read the company docs or Slack history or had conversations with members of the team, or spent any time understanding the company-internal codebase. A smart new hire isn’t that useful 5 minutes after arriving—but they are quite useful a month in! It seems like it should be possible, for example via very-long-context, to “onboard” models like we would a new human coworker. This alone would be a huge unlock.

  2. The test-time compute overhang (reasoning/error correction/system ii for longer-horizon problems.) Right now, models can basically only do short tasks: you ask them a question, and they give you an answer. But that’s extremely limiting. Most useful cognitive work humans do is longer horizon—it doesn’t just take 5 minutes, but hours, days, weeks, or months. In essence, there is a large test-time compute overhang. Think of each GPT-4 token as a word of internal monologue when you think about a problem. Each GPT-4 token is quite smart, but it can currently only really effectively use on the order of ~hundreds of tokens for chains of thought coherently (effectively as though you could only spend a few minutes of internal monologue/thinking on a problem or project). What if it could use millions of tokens to think about and work on really hard problems or bigger projects? In essence, we just need to teach the model a sort of System II outer loop that lets it reason through difficult, long-horizon projects. If we succeed at teaching this outer loop, instead of a short chatbot answer of a couple paragraphs, imagine a stream of millions of words (coming in more quickly than you can read them) as the model thinks through problems, uses tools, tries different approaches, does research, revises its work, coordinates with others, and completes big projects on its own.

  3. Using a computer. This is perhaps the most straightforward of the three. ChatGPT right now is basically like a human that sits in an isolated box that you can text. While early unhobbling improvements teach models to use individual isolated tools, I expect that with multimodal models we will soon be able to do this in one fell swoop: we will simply enable models to use a computer like a human would. That means joining your Zoom calls, researching things online, messaging and emailing people, reading shared docs, using your apps and dev tooling, and so on. Of course, for models to make the most use of this in longer-horizon loops, this will go hand-in-hand with unlocking test-time compute.

By the end of this, I expect us to get something that looks a lot like a drop-in remote worker. An agent that joins your company, is onboarded like a new human hire, messages you and colleagues on Slack and uses your softwares, makes pull requests, and that, given big projects, can do the model-equivalent of a human going away for weeks to independently complete the project. You’ll probably need somewhat better base models than GPT-4 to unlock this, but possibly not even that much better—a lot of juice is in fixing the clear and basic ways models are still hobbled.

We are on course for AGI by 2027. These AI systems will basically be able to automate basically all cognitive jobs (think: all jobs that could be done remotely). To be clear—the error bars are large. Progress could stall as we run out of data, if the algorithmic breakthroughs necessary to crash through the data wall prove harder than expected. Maybe unhobbling doesn’t go as far, and we are stuck with merely expert chatbots, rather than expert coworkers. Perhaps the decade-long trendlines break, or scaling deep learning hits a wall for real this time. (Or an algorithmic breakthrough, even simple unhobbling that unleashes the test-time compute overhang, could be a paradigm-shift, accelerating things further and leading to AGI even earlier.) In any case, we are racing through the OOMs, and it requires no esoteric beliefs, merely trend extrapolation of straight lines, to take the possibility of AGI—true AGI—by 2027 extremely seriously.

From AGI to Superintelligence: the Intelligence Explosion

AI progress won’t stop at human-level. Hundreds of millions of AGIs could automate AI research, compressing a decade of algorithmic progress (5+ OOMs) into 1 year. We would rapidly go from human-level to vastly superhuman AI systems. The power—and the peril—of superintelligence would be dramatic.

We don’t need to automate everything—just AI research. A common objection to transformative impacts of AGI is that it will be hard for AI to do everything. Look at robotics, for instance, doubters say; that will be a gnarly problem, even if AI is cognitively at the levels of PhDs. Or take automating biology R&D, which might require lots of physical lab-work and human experiments. But we don’t need robotics—we don’t need many things—for AI to automate AI research. The jobs of AI researchers and engineers at leading labs can be done fully virtually and don’t run into real-world bottlenecks in the same way (though it will still be limited by compute, which I’ll address later). And the job of an AI researcher is fairly straightforward, in the grand scheme of things: read ML literature and come up with new questions or ideas, implement experiments to test those ideas, interpret the results, and repeat. This all seems squarely in the domain where simple extrapolations of current AI capabilities could easily take us to or beyond the levels of the best humans by the end of 2027.

That 100x human speed not long after we begin to be able to automate AI research. They’ll each be able to do a year’s worth of work in a few days. The increase in research effort—compared to a few hundred puny human researchers at a leading AI lab today, working at a puny 1x human speed—will be extraordinary. This could easily dramatically accelerate existing trends of algorithmic progress, compressing a decade of advances into a year.

It’s strikingly plausible we’d go from AGI to superintelligence very quickly, perhaps in 1 year.

Whether or not you agree with the strongest form of these arguments—whether we get a <1 year intelligence explosion, or it takes a few years—it is clear: we must confront the possibility of superintelligence. The AI systems we’ll likely have by the end of this decade will be unimaginably powerful. Imagine if the technological progress of the 20th century were compressed into less than a decade. We would have gone from flying being thought a mirage, to airplanes, to a man on the moon and ICBMs in a matter of years. This is what I expect the 2030s to look like across science and technology.

How all of this plays out over the 2030s is hard to predict (and a story for another time). But one thing, at least, is clear: we will be rapidly plunged into the most extreme situation humanity has ever faced. Human-level AI systems, AGI, would be highly consequential in their own right—but in some sense, they would simply be a more efficient version of what we already know. But, very plausibly, within just a year, we would transition to much more alien systems, systems whose understanding and abilities— whose raw power—would exceed those even of humanity combined. There is a real possibility that we will lose control, as we are forced to hand off trust to AI systems during this rapid transition. More generally, everything will just start happening incredibly fast. And the world will start going insane. Suppose we had gone through the geopolitical fever-pitches and man-made perils of the 20th century in mere years; that is the sort of situation we should expect post-superintelligence. By the end of it, superintelligent AI systems will be running our military and economy. During all of this insanity, we’d have extremely scarce time to make the right decisions. The challenges will be immense. It will take everything we’ve got to make it through in one piece.

The intelligence explosion and the immediate post-superintelligence period will be one of the most volatile, tense, dangerous, and wildest periods ever in human history. And by the end of the decade, we’ll likely be in the midst of it.

The Challenges

Racing to the Trillion-Dollar Cluster

The most extraordinary techno-capital acceleration has been set in motion. As AI revenue grows rapidly, many trillions of dollars will go into GPU, datacenter, and power buildout before the end of the decade. The industrial mobilization, including growing US electricity production by 10s of percent, will be intense.

The scale of investment postulated here may seem fantastical. But both the demand-side and the supply-side seem like they could support the above trajectory. The economic returns justify the investment, the scale of expenditures is not unprecedented for a new general-purpose technology, and the industrial mobilization for power and chips is doable.

Before the decade is out, many trillions of dollars of compute clusters will have been built. The only question is whether they will be built in America. Some are rumored to be betting on building them elsewhere, especially in the Middle East. Do we really want the infrastructure for the Manhattan Project to be controlled by some capricious Middle Eastern dictatorship? The clusters that are being planned today may well be the clusters AGI and superintelligence are trained and run on, not just the “cool-big-tech-product clusters.” The national interest demands that these are built in America (or close democratic allies). Anything else creates an irreversible security risk: it risks the AGI weights getting stolen (and perhaps be shipped to China); it risks these dictatorships physically seizing the datacenters (to build and run AGI themselves) when the AGI race gets hot; or even if these threats are only wielded implicity, it puts AGI and superintelligence at unsavory dictator’s whims. America sorely regretted her energy dependence on the Middle East in the 70s, and we worked so hard to get out from under their thumbs. We cannot make the same mistake again.

Lock Down the Labs: Security for AGI

The nation’s leading AI labs treat security as an afterthought. Currently, they’re basically handing the key secrets for AGI to the CCP on a silver platter. Securing the AGI secrets and weights against the state-actor threat will be an immense effort, and we’re not on track.

All the trillions we will invest, the mobilization of American industrial might, the efforts of our brightest minds—none of that matters if China or others can simply steal the model weights (all a finished AI model is, all AGI will be, is a large file on a computer) or key algorithmic secrets (the key technical breakthroughs necessary to build AGI). America’s leading AI labs self-proclaim to be building AGI: they believe that the technology they are building will, before the decade is out, be the most powerful weapon America has ever built. But they do not treat it as such. They measure their security efforts against “random tech startups,” not “key national defense projects.” As the AGI race intensifies—as it becomes clear that superintelligence will be utterly decisive in international military competition—we will have to face the full force of foreign espionage. Currently, labs are barely able to defend against scriptkiddies, let alone have “North Koreaproof security,” let alone be ready to face the Chinese Ministry of State Security bringing its full force to bear. And this won’t just matter years in the future. Sure, who cares if GPT-4 weights are stolen—what really matters in terms of weight security is that we can secure the AGI weights down the line, so we have a few years, you might say. (Though if we’re building AGI in 2027, we really have to get moving!) But the AI labs are developing the algorithmic secrets—the key technical breakthroughs, the blueprints so to speak—for the AGI right now (in particular, the RL/self-play/synthetic data/etc “next paradigm” after LLMs to get past the data wall). AGI-level security for algorithmic secrets is necessary years before AGIlevel security for weights. These algorithmic breakthroughs will matter more than a 10x or 100x larger cluster in a few years—this is a much bigger deal than export controls on compute, which the USG has been (presciently!) intensely pursuing. Right now, you needn’t even mount a dramatic espionage operation to steal these secrets: just go to any SF party or look through the office windows.

There are two key assets we must protect: model weights (especially as we get close to AGI, but which takes years of preparation and practice to get right) and algorithmic secrets (starting yesterday).

It’s easy to underrate how important an edge algorithmic secrets will be—because up until ~a couple years ago, everything was published. The basic idea was out there: scale up Transformers on internet text. Many algorithmic details and efficiencies were out there: Chinchilla scaling laws, MoE, etc.Thus, open source models today are pretty good, and a bunch of companies have pretty good models (mostly depending on how much $$$ they raised and how big their clusters are). But this will likely change fairly dramatically in the next couple years. Basically all of frontier algorithmic progress happens at labs these days (academia is surprisingly irrelevant), and the leading labs have stopped publishing their advances. We should expect far more divergence ahead: between labs, between countries, and between the proprietary frontier and open source models. A few American labs will be way ahead—a moat worth 10x, 100x, or more, way more than, say, 7nm vs. 3nm chips—unless they instantly leak the algorithmic secrets. I sometimes joke that AI lab algorithmic advances are not shared with the American research community, but they are being shared with the Chinese research community! Put simply, I think failing to protect algorithmic secrets is probably the most likely way in which China is able to stay competitive in the AGI race.

Once China begins to truly understand the import of AGI, we should expect the full force of their espionage efforts to come to bear; think billions of dollars invested, thousands of employees, and extreme measures (e.g., special operations strike teams) dedicated to infiltrating American AGI efforts. What will security for AGI and superintelligence require? In short, this will only be possible with government help. Microsoft, for example, is regularly hacked by state actors (e.g., Russian hackers recently stole Microsoft executives’ emails, as well as government emails Microsoft hosts). A high-level security expert working in the field estimated that even with a complete private crash course, China would still likely be able to exfiltrate the AGI weights if it was their #1 priority—the only way to get this probability to the single digits would require, more or less, a government project. While the government does not have a perfect track record on security themselves, they’re the only ones who have the infrastructure, know-how, and competencies to protect nationaldefense-level secrets. Basic stuff like the authority to subject employees to intense vetting; threaten imprisonment for leaking secrets; physical security for datacenters; and the vast know-how of places like the NSA and the people behind the security clearances (private companies simply don’t have the expertise on state-actor attacks.)

We’re developing the most powerful weapon mankind has ever created. The algorithmic secrets we are developing, right now, are literally the nation’s most important national defense secrets—the secrets that will be at the foundation of the US and her allies’ economic and military predominance by the end of the decade, the secrets that will determine whether we have the requisite lead to get AI safety right, the secrets that will determine the outcome of WWIII, the secrets that will determine the future of the free world. And yet AI lab security is probably worse than a random defense contractor making bolts. It’s madness. Basically nothing else we do—on national competition, and on AI safety—will matter if we don’t fix this, soon.

Superalignment

Reliably controlling AI systems much smarter than we are is an unsolved technical problem. And while it is a solvable problem, things could very easily go off the rails during a rapid intelligence explosion. Managing this will be extremely tense; failure could easily be catastrophic.

By the time the decade is out, we’ll have billions of vastly superhuman AI agents running around. These superhuman AI agents will be capable of extremely complex and creative behavior; we will have no hope of following along. We’ll be like first graders trying to supervise with multiple doctorates. In essence, we face a problem of handing off trust. By the end of the intelligence explosion, we won’t have any hope of understanding what our billion superintelligences are doing (except as they might choose to explain to us, like they might to a child). And we don’t yet have the technical ability to reliably guarantee even basic side constraints for these systems, like “don’t lie” or “follow the law” or “don’t try to exfiltrate your server”.

Simply put, without a very concerted effort, we won’t be able to guarantee that superintelligence won’t go rogue (and this is acknowledged by many leaders in the field).Yes, it may all be fine by default. But we simply don’t know yet. Especially once future AI systems aren’t just trained with imitation learning, but large-scale, long-horizon RL (reinforcement learning), they will acquire unpredictable behaviors of their own, shaped by a trial-and-error process (for example, they may learn to lie or seek power, simply because these are successful strategies in the real world!). The stakes will be high enough that hoping for the best simply isn’t a good enough answer on alignment.

The core technical problem of superalignment is simple: how do we control AI systems (much) smarter than us? RLHF will predictably break down as AI systems get smarter, and we will face fundamentally new and qualitatively different technical challenges. Imagine, for example, a superhuman AI system generating a million lines of code in a new programming language it invented. If you asked a human rater in an RLHF procedure, “does this code contain any security backdoors?” they simply wouldn’t know. They wouldn’t be able to rate the output as good or bad, safe or unsafe, and so we wouldn’t be able to reinforce good behaviors and penalize bad behaviors with RLHF.

The superalignment problem being unsolved means that we simply won’t have the ability to ensure even these basic side constraints for these superintelligence systems, like “will they reliably follow my instructions?” or “will they honestly answer my questions?” or “will they not deceive humans?”. People often associate alignment with some complicated questions about human values, or jump to political controversies, but deciding on what behaviors and values to instill in the model, while important, is a separate problem. The primary problem is that for whatever you want to instill the model (including ensuring very basic things, like “follow the law”!) we don’t yet know how to do that for the very powerful AI systems we are building.

One example that’s very salient to me: we may well bootstrap our way to human-level or somewhat-superhuman AGI with systems that reason via chains of thoughts, i.e. via English tokens. This is extraordinarily helpful, because it means the models “think out loud” letting us catch malign behavior (e.g., if it’s scheming against us). But surely having AI systems think in tokens is not the most efficient means to do it, surely there’s something much better that does all of this thinking via internal states—and so the model by the end of the intelligence explosion will almost certainly not think out loud, i.e. will have completely uninterpretable reasoning.

I think we can harvest wins across a number of empirical bets, which I’ll describe below, to align somewhat-superhuman systems. Then, if we’re confident we can trust these systems, we’ll need to use these somewhat-superhuman systems to automate alignment research—alongside the automation of AI research in general, during the intelligence explosion—to figure out how to solve alignment to go the rest of the way.

Here are some of the main research bets I see for crossing the gap between human-level and somewhat-superhuman systems.

Evaluation is easier than generation. We get some of the way “for free,” because it’s easier for us to evaluate outputs (especially for egregious misbehaviors) than it is to generate them ourselves.

Scalable oversight. We can use AI assistants to help humans supervise other AI systems—the human-AI team being able to extend supervision farther than the human could alone.

Generalization. Even with scalable oversight, we won’t be able to supervise AI systems on really hard problems, problems beyond human comprehension. However, we can study: how will the AI systems generalize from human supervision on easy problems (that we do understand and can supervise) to behave on the hard problems (that we can’t understand and can no longer supervise)?

Interpretability. One intuitively-attractive way we’d hope to verify and trust that our AI systems are aligned is if we could understand what they’re thinking! For example, if we’re worried that AI systems are deceiving us or conspiring against us, access to their internal reasoning should help us detect that.

Adversarial testing and measurements. Along the way, it’s going to be critical to stress test the alignment of our systems at every step—our goal should be to encounter every failure mode in the lab before we encounter it in the wild. This will require substantially advancing techniques for automated red-teaming.

The intelligence explosion will be more like running a war than launching a product. We’re not on track for superdefense, for an airgapped cluster or any of that; I’m not sure we would even realize if a model self-exfiltrated. We’re not on track for a sane chain of command to make any of these insanely high-stakes decisions, to insist on the very-high-confidence appropriate for superintelligence, to make the hard decisions to take extra time before launching the next training run to get safety right or dedicate a large majority of compute to alignment research, to recognize danger ahead and avert it rather than crashing right into it. Right now, no lab has demonstrated much of a willingness to make any costly tradeoffs to get safety right (we get lots of safety committees, yes, but those are pretty meaningless). By default, we’ll probably stumble into the intelligence explosion and have gone through a few OOMs before people even realize what we’ve gotten into. We’re counting way too much on luck here.

The Free World Must Prevail

Superintelligence will give a decisive economic and military advantage. China isn’t at all out of the game yet. In the race to AGI, the free world’s very survival will be at stake. Can we maintain our preeminence over the authoritarian powers? And will we manage to avoid self-destruction along the way?

Our generation too easily takes for granted that we live in peace and freedom. And those who herald the age of AGI in SF too often ignore the elephant in the room: superintelligence is a matter of national security, and the United States must win.

If and when the CCP wakes up to AGI, we should expect extraordinary efforts on the part of the CCP to compete. And I think there’s a pretty clear path for China to be in the game: outbuild the US and steal the algorithms.

The free world must prevail over the authoritarian powers in this race. We owe our peace and freedom to American economic and military preeminence. Perhaps even empowered with superintelligence, the CCP will behave responsibly on the international stage, leaving each to their own. But the history of dictators of their ilk is not pretty. If America and her allies fail to win this race, we risk it all.

The US has a lead. We just have to keep it. And we’re screwing that up right now. Most of all, we must rapidly and radically lock down the AI labs, before we leak key AGI breakthroughs in the next 12-24 months (or the AGI weights themselves). We must build the compute clusters in the US, not in dictatorships that offer easy money. And yes, American AI labs have a duty to work with the intelligence community and the military. America’s lead on AGI won’t secure peace and freedom by just building the best AI girlfriend apps. It’s not pretty—but we must build AI for American defense.

We are already on course for the most combustive international situation in decades. Putin is on the march in Eastern Europe. The Middle East is on fire. The CCP views taking Taiwan as its destiny. Now add in the race to AGI. Add in a century’s worth of technological breakthroughs compressed into years post-superintelligence. It will be the one of most unstable international situations ever seen—and at least initially, the incentives for first-strikes will be tremendous.

There’s already an eerie convergence of AGI timelines (~2027?) and Taiwan watchers’ Taiwan invasion timelines (China ready to invade Taiwan by 2027?)—a convergence that will surely only heighten as the world wakes up to AGI. (Imagine if in 1960, the vast majority of the world’s uranium deposits were somehow concentrated in Berlin!) It seems to me that there is a real chance that the AGI endgame plays out with the backdrop of world war. Then all bets are off.

The Project

As the race to AGI intensifies, the national security state will get involved. The USG will wake from its slumber, and by 27/28 we’ll get some form of government AGI project. No startup can handle superintelligence. Somewhere in a SCIF, the endgame will be on.

Many plans for “AI governance” are put forth these days, from licensing frontier AI systems to safety standards to a public cloud with a few hundred million in compute for academics. These seem well-intentioned—but to me, it seems like they are making a category error. I find it an insane proposition that the US government will let a random SF startup develop superintelligence. Imagine if we had developed atomic bombs by letting Uber just improvise.

Like many scientists before us, the great minds of San Francisco hope that they can control the destiny of the demon they are birthing. Right now, they still can; for they are among the few with situational awareness, who understand what they are building. But in the next few years, the world will wake up. So too will the national security state. History will make a triumphant return. As in many times before—Covid, WWII—it will seem as though the United States is asleep at the wheel—before, all at once, the government shifts into gear in the most extraordinary fashion. There will be a moment—in just a few years, just a couple more “2023-level” leaps in model capabilities and AI discourse— where it will be clear: we are on the cusp of AGI, and superintelligence shortly thereafter. While there’s a lot of flux in the exact mechanics, one way or another, the USG will be at the helm; the leading labs will (“voluntarily”) merge; Congress will appropriate trillions for chips and power; a coalition of democracies formed.

Slowly at first, then all at once, it will become clear: this is happening, things are going to get wild, this is the most important challenge for the national security of the United States since the invention of the atomic bomb. In one form or another, the national security state will get very heavily involved. The Project will be the necessary, indeed the only plausible, response.

Whether nominally private or not, the AGI project will need to be, will be, integrally a defense project, and it will require extremely close cooperation with the national security state. We will need a sane chain of command—along with all the other processes and safeguards that necessarily come with responsibly wielding what will be comparable to a WMD—and it’ll require the government to do so. In some sense, this is simply a Burkean argument: the institutions, constitutions, laws, courts, checks and balances, norms and common dedication to the liberal democratic order (e.g., generals refusing to follow illegal orders), and so on that check the power of the government have withstood the test of hundreds of years. Special AI lab governance structures, meanwhile, collapsed the first time they were tested. The US military could already kill basically every civilian in the United States, or seize power, if it wanted to—and the way we keep government power over nuclear weapons in check is not through lots of private companies with their own nuclear arsenals. There’s only one chain of command and set of institutions that has proven itself up to this task.

The initial development of superintelligence will be dominated by the national security exigency to survive and stabilize an incredibly volatile period. And the military uses of superintelligence will remain reserved for the government, and safety norms will be enforced. But once the initial peril has passed, the natural path is for the companies involved in the national consortium (and others) to privately pursue civilian applications. Even in worlds with The Project, a private, pluralistic, market-based, flourishing ecosystem of civilian applications of superintelligence will have its day.

The intelligence explosion and its immediate aftermath will bring forth one of the most volatile and tense situations mankind has ever faced. Our generation is not used to this. But in this initial period, the task at hand will not be to build cool products. It will be to somehow, desperately, make it through this period.

Ultimately, my main claim here is descriptive: whether we like it or not, superintelligence won’t look like an SF startup, and in some way will be primarily in the domain of national security.

And so by 27/28, the endgame will be on. By 28/29 the intelligence explosion will be underway; by 2030, we will have summoned superintelligence, in all its power and might. Whoever they put in charge of The Project is going to have a hell of a task: to build AGI, and to build it fast; to put the American economy on wartime footing to make hundreds of millions of GPUs; to lock it all down, weed out the spies, and fend off all-out attacks by the CCP; to somehow manage a hundred million AGIs furiously automating AI research, making a decade’s leaps in a year, and soon producing AI systems vastly smarter than the smartest humans; to somehow keep things together enough that this doesn’t go off the rails and produce rogue superintelligence that tries to seize control from its human overseers; to use those superintelligences to develop whatever new technologies will be necessary to stabilize the situation and stay ahead of adversaries, rapidly remaking US forces to integrate those; all while navigating what will likely be the tensest international situation ever seen. They better be good, I’ll say that.

Parting Thoughts

As I see it, the smartest people in the space have converged on a different perspective, a third way, one I will dub AGI Realism. The core tenets are simple:

  1. Superintelligence is a matter of national security. We are rapidly building machines smarter than the smartest humans. This is not another cool Silicon Valley boom; this isn’t some random community of coders writing an innocent open source software package; this isn’t fun and games. Superintelligence is going to be wild; it will be the most powerful weapon mankind has ever built. And for any of us involved, it’ll be the most important thing we ever do

  2. America must lead. The torch of liberty will not survive Xi getting AGI first. (And, realistically, American leadership is the only path to safe AGI, too.) That means we can’t simply “pause”; it means we need to rapidly scale up US power production to build the AGI clusters in the US. But it also means amateur startup security delivering the nuclear secrets to the CCP won’t cut it anymore, and it means the core AGI infrastructure must be controlled by America, not some dictator in the Middle East. American AI labs must put the national interest first.

  3. We need to not screw it up. Recognizing the power of superintelligence also means recognizing its peril. There are very real safety risks; very real risks this all goes awry—whether it be because mankind uses the destructive power brought forth for our mutual annihilation, or because, yes, the alien species we’re summoning is one we cannot yet fully control. These are manageable—but improvising won’t cut it. Navigating these perils will require good people bringing a level of seriousness to the table that has not yet been offered.

But the scariest realization is that there is no crack team coming to handle this. As a kid you have this glorified view of the world, that when things get real there are the heroic scientists, the uber-competent military men, the calm leaders who are on it, who will save the day. It is not so. The world is incredibly small; when the facade comes off, it’s usually just a few folks behind the scenes who are the live players, who are desperately trying to keep things from falling apart. Right now, there’s perhaps a few hundred people in the world who realize what’s about to hit us, who understand just how crazy things are about to get, who have situational awareness. I probably either personally know or am one degree of separation from everyone who could plausibly run The Project. The few folks behind the scenes who are desperately trying to keep things from falling apart are you and your buddies and their buddies. That’s it. That’s all there is. Someday it will be out of our hands. But right now, at least for the next few years of midgame, the fate of the world rests on these people.

Will the free world prevail?

Will we tame superintelligence, or will it tame us?

Will humanity skirt self-destruction once more?

The stakes are no less.

These are great and honorable people. But they are just people. Soon, the AIs will be running the world, but we’re in for one last rodeo. May their final stewardship bring honor to mankind.