Will AGI be achieved in the next 5 years?
➕
Plus
93
Ṁ7918
2029
38%
chance

AGI existence is determined by general consensus

Resolves to YES if general consensus of those who interact with the potential AGI is that it is AGI.

Resolves to NO if after 5 years (January 19 2029) it has not resolved to YES.

Get Ṁ1,000 play money
Sort by:

The close date should be extended to 2029

Now Claude 3 beats the average human with an IQ of 101. It has a ridiculous context window and I can put an entire codebase into it and it outputs changes to any line like magic.

Are we still going to pretend this hasn't resolved YES yet?

I never bet on these markets because GPT-4 wasn't AGI, now the 1m context Gemini isn't, and Sora isn't, so what ever will be?

@SteveSokolowski Gemini certainly isn't.

I asked it to write a blog post in my style, after inputting my entire blog there, and it failed miserably.

predicts NO

Can this be unclosed? @kalb

predicts NO

@Leminspector Oh, looks like he just accidentally left the default one week from creation close time. Fixed.

The question would benefit from a rephrasing like "will AGI be achieved by January 2029?"

One meaning stays the same over time, the other drifts.

Why is the close date in Jan 2024?

predicts YES

Does anyone know what Manifold's policy on this is? Will the moderators manually extend the close time if it clearly matches the title/description of the question?

Is AGI in this case defined as at-human intelligence or above-human intelligence?

predicts NO

@NateIO AGI is defined as whatever the general population believes is AGI. In the next 5 years, if the general consensus of the population believes an AI is AGI, this market will resolve to YES. They would have to use that specific term.

For example, GPT-4 would not resolve the market to YES because while some people were saying it was AGI, it never reached close to general consensus.