Caveat aleator: the definition of AGI for this market is a work in progress -- ask clarifying questions!

AI keeps getting capabilities that I previously would've thought would require AGI. Solving FrontierMath problems, explaining jokes, generating art and poetry. And this isn't new. We once presumed that playing grandmaster-level chess would require AGI. I'd like to have a market for predicting when we'll reach truly human-level intelligence -- mentally, not physically -- that won't accidentally resolve YES on a technicality.

But I also want to allow for the perfectly real possibility that people like Eric Drexler are right and that we get AGI without the world necessarily turning upside-down.

I think the right balance, what best captures the spirit of the question, is to focus on Leopold Aschenbrenner's concept of drop-in remote workers.

However, it's possible that that definition is also flawed. So, DO NOT TRADE IN THIS MARKET YET. Ask more clarifying questions until this FAQ is fleshed out enough that you're satisfied that we're predicting the right thing.

FAQ

1. Will this resolve the same as other AGI markets on Manifold?

It probably will but the point of this market is to avoid getting backed into a corner on a definition of AGI that violates the spirit of the question.

2. Does passing a long, informed, adversarial Turing Test count?

Not necessarily. Matthew Barnett on Metaculus has a lengthy definition of this kind of Turning test. I believe it's even stricter than the Longbets version. Roughly the idea is to have human foils who are PhD-level experts in at least one STEM field, judges who are PhD-level experts in AI, with each interview at least 2 hours and with everyone trying their hardest to act human and distinguish humans from AIs.

I think it's likely that a test like that can work (see discussion in the comments for ideas for making it robust enough). But ultimately for this market we need to be robust to the possibility that something that's spiritually AGI fails a Turing test for dumb reasons like being unwilling to lie or being too smart. And we need to be robust to non-AGI passing by being able to stay coherent for some hours during the test without being able to do planning and problem-solving over the course of days/weeks/months like humans can.

Whatever weight we put on a Turing test, we have to ensure that the AI is not passing or failing based on anything unrelated to its actual intelligence/capability.

3. Does the AGI have to be publicly available?

No but we won't just believe Sam Altman or whoever and in fact we'll err on the side of disbelieving. (This needs to be pinned down further.)

4. What if running the AGI is obscenely slow and expensive?

The focus is on human-level AI so if it's more expensive than hiring a human at market rates and slower than a human, that doesn't count.

5. What if it's cheaper but slower than a human, or more expensive but faster?

Again, we'll use the drop-in remote worker concept here. I don't know the threshold yet but it needs to commonly make sense to use an artificial worker rather than hire a human. Being sufficiently cheaper despite being slower or being sufficiently faster despite being pricier could count.

6. If we get AGI in, say, 2027, does 2028 also resolve YES?

No, we're predicting the exact year. This is a probability density function (pdf), not a cumulative distribution function (cdf).

Ask clarifying questions before betting! I'll add them to the FAQ.

Note that I'm betting in this market myself and am committing to make the resolution fair. Meaning I'll be transparent about my reasoning and if there are good faith objections, I'll hear them out, we'll discuss, and I'll outsource the final decision if needed. But, again, note the evolving FAQ. The expectation is that bettors ask clarifying questions before betting, to minimize the chances of it coming down to a judgment call.

Related Markets

Scratch area for auto-generated AI updates

Don't believe what magically appears down here; I'll add clarifications to the FAQ.

Update 2025-04-02 (PST) (AI summary of creator comment): Definition of AGI (for market resolution)
- Remote worker preference: The AI must be preferred by anyone hiring a remote worker based solely on its ability to complete tasks, without considering legal, ethical, or PR concerns.
- Work performance criterion: In terms of actual work output, the AI should outperform a human either by being vastly more cost-effective (making hiring a rare, expensive superstar unnecessary) or by avoiding the low performance typical of a barely-literate human.
- Coverage of the Pareto frontier: To count as AGI for this market, the AI should cover almost all of the Pareto frontier in task capability.

Update 2025-04-09 (PST) (AI summary of creator comment): Update from creator
- The AGI definition will be based on the criteria outlined in the provided DeepMind paper.
- A key resolution parameter will be the performance threshold, with options being competent, expert, or virtuoso.
- The final market outcome will depend on which of these thresholds is selected.