When I wrote about AI 2027, I called its race scenario plausible and frightening. It showed AI labs using increasingly capable agents to accelerate their own research until human oversight could no longer keep up. AI 2040 is the missing second half: if continuing the race ends badly, what would stopping it actually require?

Thomas Larsen and the other authors at the AI Futures Project are explicit that Plan A is a recommendation rather than their best prediction:

Plan A is our positive vision for how humanity can avoid AI-driven existential catastrophe and reach a flourishing future. It’s informed by conversations with experts at major U.S. frontier AI companies, direct experience at OpenAI, tabletop exercises, and discussions with policymakers, national security experts, and AI policy leaders. We recommend an international deal to avoid a dangerous race to superintelligence. The deal involves total research transparency for AI R&D, which allows the nations of the world to understand what’s happening and enforce guardrails. The result is multiple companies across multiple countries scaling slowly and safely together towards superintelligence, instead of racing each other in secrecy.

The recommendation changes how the scenario should be read. The earlier scenario asked readers to choose between slowing down and racing ahead after the machinery was already in motion. Plan A works backward from the slowdown branch and makes the political machinery concrete: a verifiable U.S.–China agreement, public AI research, more companies brought to the frontier, and a threat to destroy compute if either side breaks the deal.

The title can make the new scenario sound like the authors have simply pushed their forecast thirteen years into the future. They have not:

The timeline of this scenario is:

• In 2029, the US and China agree to avoid a reckless race to superintelligence.
• In 2030, we would have fully automated AI R&D, leading to superintelligence by the end of the year. Thanks to the deal, we avoid this.
• Between 2030 and 2035, we scale within the human range, to AIs that are roughly as capable as top human experts.
• In 2035, we pause at top-human-expert level AI in order to maintain human control.
• In 2040, we unpause and scale to superintelligence. (Hence the title: AI 2040)

In our previous scenario, AI 2027, AI fully automated the process of building smarter AIs in 2027, leading to an intelligence explosion and superintelligence within the year. The two differences in this scenario are (1) the default timeline is now 2030, and (2) thanks to governance actions, generally-superhuman AIs first appear in 2040.

The baseline moves from 2027 to 2030; governance creates the remaining decade. In this scenario, reaching 2040 is an achievement produced by intervention. It also moves the decisive problem from whether labs can align a superintelligence to whether governments can coordinate before the labs build one. The proposal is ambitious enough to invite skepticism. That is exactly why the authors render it in such detail.

We think most AI policy proposals fall apart under scenario scrutiny—that is, if you try to write down a detailed and plausible scenario in which that proposal succeeds, you will find it difficult to do so, and you will realize the plan is less likely to work than it seemed, or has more unpleasant side-effects than its proponents acknowledged.

Perhaps that’s why scenario scrutiny is so rare in AI policy. Everyone wants to say that their own favorite policies will have great consequences and that the policies of their rivals will have terrible consequences. Applying scenario scrutiny to their own favorite policies might surface uncomfortable issues with them; meanwhile, applying scenario scrutiny to their rival’s policies is a lot of work for little rhetorical gain.

We think the discourse would be improved if more AI policy proposals were subjected to scenario scrutiny. So we’re starting with our own, even though this opens us up to criticism. We hope critics will judge us against the existing state-of-the-art for plans to navigate the AI transition (if they can find any) and not against some hazy but pleasant fantasy where no one has to make any hard choices yet everything will probably be fine.

You don’t have to accept every date or mechanism to take this seriously. Read Plan A as a stress test: examine where its chain of coordination might break, then ask whether your preferred alternative survives the same level of scrutiny. Dismissing the scenario because it is speculative only leaves us with less explicit plans for navigating the same risk.

Is this about design? No. But it is about humanity’s future.