The AI 2040 scenario is the follow-up to AI 2027 from an overlapping group of researchers, and it proposes something different from the earlier document: not just a prediction of what might happen, but a recommended path for what should happen instead. If you read AI 2027 and came away worried about a fast, uncontrolled timeline, the AI 2040 scenario is worth understanding as the same research community’s answer to that worry.
AI 2040, formally titled “Plan A,” was written by Thomas Larsen, Romeo Dean, Brendan Halstead, Eli Lifland, Ryan Greenblatt, and Daniel Kokotajlo of the AI Futures Project, several of whom also worked on AI 2027. Where AI 2027 described a fast, largely uncontrolled race toward advanced AI, AI 2040 lays out a slower, deliberately governed alternative timeline running through 2040, along with the international coordination the authors believe would be needed to get there.
This guide breaks down what the AI 2040 scenario actually proposes, how it differs from AI 2027, and what practical career takeaways make sense whether this specific proposal is adopted, ignored, or something in between.
Treating this as just a sequel to a document you already understand skips the more important shift: AI 2040 is prescriptive rather than purely predictive, and that distinction changes how you should weigh its career implications.
The Short Answer
The AI 2040 scenario, or “Plan A,” proposes that the US and China agree by 2029 to avoid racing toward superintelligent AI, then scale capabilities slowly and safely together through 2035, pause deliberately to maintain human control, and only resume toward more advanced systems by 2040. It is a recommended governance path, not a forecast of what the authors expect to actually happen by default.
The most useful way to read it is as a stress test of what a genuinely safer alternative would require, including real sacrifices and coordination problems, rather than as a comforting prediction that a slow, careful timeline is already underway.
For your career specifically, the honest takeaway is that AI 2040 describes a possible future, not a current plan already in motion, so it should inform your thinking without becoming the basis for a major decision on its own.
What the AI 2040 Scenario Actually Proposes
| Timeframe | Milestone in the Plan A Scenario | Purpose |
|---|---|---|
| 2029 | US and China agree to avoid racing toward superintelligence | Establish coordination before capabilities accelerate |
| 2030-2035 | AI capabilities scaled to match top human experts, slowly and jointly | Maintain parity and verification between parties |
| 2035 | Intentional pause in further scaling | Preserve human control and oversight |
| 2040 | Scaling toward more advanced systems resumes | Continue development under established safeguards |
The table above reflects the general shape of the proposal, moving from an initial coordination agreement toward a long, deliberately paced scaling period. The authors describe the default trajectory, meaning what happens without this kind of intervention, as fully automated AI research and development arriving by around 2030, followed by a rapid intelligence explosion within months. The AI 2040 scenario is explicitly an attempt to avoid that default path.
The proposal centers on international coordination rather than a single company or country acting alone. It calls for what the authors describe as total research transparency around AI development, verification mechanisms between competing nations, and a framework sometimes described as mutually assured compute destruction, meaning capacity is deliberately constrained in a way both sides can verify.
A deliberate pause around 2035 is a central feature of the plan. Rather than continuing to scale capability as fast as possible, the scenario describes intentionally slowing down at that point specifically to preserve human oversight and control before any further advancement toward superintelligence resumes closer to 2040.
How AI 2040 Differs From AI 2027
The most important shift between the two documents is the default timeline itself. AI 2027 was built around a scenario reaching a critical point by September 2027. AI 2040 moves the assumed default timeline for fully automated AI research out to around 2030, reflecting the authors’ own updated uncertainty about how fast this technology will actually progress.
The second major difference is purpose. AI 2027 reads as a narrative prediction of one plausible path, largely uncontrolled, that the authors consider genuinely possible. AI 2040, by contrast, is explicitly normative: the authors state they want to advocate for something that is “actually good,” rather than simply describing a lesser-evil outcome among bad options.
This means AI 2040 should not be read as the same research group predicting a better, calmer version of the same future is already likely. It is closer to a policy proposal wrapped in scenario-planning format, aimed at people who might actually influence whether coordination like this happens.
What This Means If You Are Job Searching Right Now
Both documents describe significant white-collar disruption as a feature of the underlying technology trend, not something unique to either the fast or slow timeline. The AI 2040 scenario specifically describes millions of AI agent copies working around the clock at superhuman speeds by 2027, with white-collar professions experiencing disruption similar to what the earlier document described for software engineering specifically.
The practical difference for your own planning is timing and uncertainty, not direction. A slower, more coordinated timeline gives individuals, employers, and policymakers more time to adapt, but it does not change the underlying claim that significant disruption to knowledge work is a realistic possibility worth preparing for regardless of which specific timeline plays out. The AI 2040 scenario does not remove that underlying uncertainty, it simply proposes a different, more deliberate path through it.
This is a reasonable moment to revisit concrete habits that actually strengthen a career regardless of how AI timelines unfold, since skills like clear communication, demonstrated judgment, and adaptability hold up whether disruption arrives on a 2027 timeline, a 2040 timeline, or something in between.
Checking scenario claims like these against real, currently measured labor market data, such as the Stanford Digital Economy Lab’s Canaries Dashboard, remains the most useful way to calibrate how seriously to take any single claim from either document. That dashboard has found slower early-career employment growth in high AI-exposure occupations, a real and current data point rather than a speculative one.
Common Mistakes People Make When Reacting to the AI 2040 Scenario
- Assuming a slower proposed timeline means slower actual disruption. AI 2040 describes a governance path, not a guarantee that job market effects will also arrive more gradually.
- Treating it as confirmation the fast AI 2027 timeline was wrong. The two documents serve different purposes, one descriptive and one prescriptive, rather than one simply correcting the other.
- Ignoring it because international coordination sounds unlikely. Even an unlikely proposal can reveal what a genuinely safer path would require, which is useful information on its own.
- Making a major career decision based on either scenario alone. Both documents are speculative inputs worth weighing alongside real labor market data, not standalone roadmaps.
Any single one of these mistakes is understandable, since scenario-planning documents like this are genuinely unusual to encounter, but together they explain why reactions to the AI 2040 scenario tend to cluster at unhelpful extremes rather than a measured, evidence-informed middle ground.
Frequently Asked Questions About the AI 2040 Scenario
Is AI 2040 a prediction of what will actually happen?
Not exactly. The authors present it as a recommended path, sometimes called “Plan A,” rather than a forecast of the most likely outcome. It describes what a genuinely safer alternative to an uncontrolled race would require, including real coordination and sacrifice, more than what they expect to happen by default.
How is AI 2040 different from AI 2027 in terms of timeline?
AI 2027 centered its scenario around a critical point in 2027. AI 2040 shifts the assumed default timeline for fully automated AI research and development out to around 2030, and proposes a deliberately paced path running through 2040 instead of a fast, uncontrolled one.
Does AI 2040 mean job disruption will happen more slowly?
Not necessarily. The scenario describes the same underlying disruption risk to white-collar work as AI 2027, just under a governance framework intended to give society more time to adapt and retain oversight. A slower governed timeline is not the same claim as slower actual disruption to any specific job category.
Should I change my career plans based on this scenario?
A single scenario document, whether AI 2027 or AI 2040, is a weak basis for a major career pivot on its own. Using either as one input alongside real labor market data, and building genuinely durable skills, is a more balanced response than treating either document as a confirmed outcome.
Who actually wrote AI 2040 and are they credible sources?
It was written by researchers at the AI Futures Project, several of whom also worked on AI 2027, with real backgrounds in AI safety and policy research. That does not make the scenario a certainty, but it does mean it reflects informed, serious analysis rather than casual speculation.
The Bottom Line
The AI 2040 scenario is best understood as a proposed alternative to an uncontrolled AI race, not a comforting update suggesting things will move more slowly by default. It shares much of AI 2027’s underlying concern about disruption to white-collar work, while offering a specific, if difficult, coordination path meant to preserve human oversight along the way.
For your own career, the practical response holds steady across both documents: build durable, transferable skills, pay attention to real measured labor market data rather than only scenario narratives, and avoid treating any single speculative document, however well-researched, as a confirmed roadmap for your own decisions.
Anonymous and approximate. We store one number for this article and nothing about you.
Email digest
We also run a layoff tracker, built from company filings and official notices. A plain email summary of what changed on these trackers: headline numbers, the largest new entries, and links to the sources. You confirm your address by clicking a link we email you, and one click unsubscribes. Already subscribed? Use this form to change what you get. The boxes replace what you had.
Our mail provider records whether you open an email and which links you follow, including the confirmation email, which is measured before you have agreed to anything. Unsubscribing stops the sending and the recording together.
You can read every past edition first. Each one is kept permanently, exactly as it was sent.
Privacy note: what we store and how to erase it
What we store: your email address, the choices above, and timestamps (signed up, confirmed, last sent). Nothing else about you.
What our mail provider records: whether you opened an email and which links you followed, tied to your address. That is Brevo, the service that delivers these emails, and it works by adding a small invisible image and by routing links through its own address. We read it to see which sections people use. Unsubscribing stops the sending and the measuring at the same time, and erasing your row removes it from our side.
The confirmation email is measured too: the same invisible image is in the one message that asks your permission. If you open it, Brevo records that against your address before you have agreed to anything, and it records it even if you then never confirm. We would exempt that message if we could. Brevo offers no way to switch its measuring off for a single email, and the one setting that exists covers every email we send or none of them. If you would rather not be measured at all, do not open it. Nothing else is ever sent to an unconfirmed address, and the signup is deleted from our side after 30 days.
About the links: links in the digest pass through a counter on this site that adds 1 to a total for that link and sends you straight on. It records no identifier, no IP address and no browser details, so it counts how many times a link was followed and can never say who followed it.
What it is used for: sending you exactly the emails you ticked, nothing else. The address is never shared, sold, or used for any other purpose.
How to erase it: click the unsubscribe link in any email. That stops all sending immediately, and unsubscribed addresses, along with signups that were never confirmed, are hard-deleted automatically after 30 days. You can also ask via the contact page.