Will non-human intelligence be confirmed before artificial general intelligence?
19
แน€479
2040
18%
chance

Which type of intelligence do you think will be confirmed by humans first?

This market will resolve to the type of superintelligence that is confirmed to exist first, on the day the threshold of clear and convincing evidence or either exists.

  • A non-human intelligence is an entity that is smarter than every single individual human, and which was not created by any human alive at the time the market opened. The market will resolve to YES if this type of intelligence is confirmed first.

  • An artificial general intelligence is an entity that is smarter than every single individual human, and which was created by one or more humans alive at the time the market opened. The market will resolve to NO if this type of intelligence is confirmed first.

This market uses a different standard of evidence than other markets on these topics. The three most commonly used legal standards are:

  • Preponderance of the evidence

  • Clear and convincing evidence

  • Beyond a reasonable doubt

One could argue that if a civil trial were to be held at the time the market opened, a preponderance of the evidence already exists for the existence of non-human intelligence, as 40 credible high-level witnesses have directly testified under oath to that fact and the behavior of officials in Congress provides additional circumstantial evidence. However, a preponderance is not this market's standard. Beyond a reasonable doubt is also not the standard, as people will undoubtedly argue over these things indefinitely.

Clear and convincing evidence would be satisfied with the following as an example (among other things):

  • OpenAI declares the creation of AGI, switching to their non-profit mode

  • The Ray Kurzweil longbet is resolved such that its very hard Turing Test is passed by the AGI

  • A country passes a law granting personhood rights to computer software that existed at the time the law was passed

  • NASA's UAP office successfully trains its model and announces that its model has inferred that UFOs are directed by non-human intelligence

  • The official stance of the US Department of Defense becomes that non-human intelligence exists

  • A whistleblower of unimpeachable character produces a UFO video taken before the advent of AI and can definitively prove the chain of custody

Note that these things would not be absolute proof, but would mark a convincing shift in the world's views in favor of these one of the topics.

If the AGI infers the existence of non-human intelligence, but we don't know the AGI existed and was responsible for that proof, this market will still resolve to YES.

If an event is a close call, the default action will always be to keep the market open.

Get แน€1,000 play money
Sort by:

What if an AGI time traveled from the time that the market closed to current time. Does it count as Yes or No?

@633b If it were created by a human alive before now, then it would be YES; otherwise NO.

Why is "created by any human alive at the time the market opened" relevant to the definitions used???

What if a child born the day after you made this market becomes the programmer to develop a superintelligent computer program in their basement?

According to your definitions this is a "non-human intelligence" rather than an "artificial general intelligence", but that seems ludicrous.

Can you reword it or expand on what you mean?
Shouldn't it just be whether it was created by a human?
Or some definition of alien intelligence vs terrestial programming?

@MatthewLeong Grusch testified that one of the working theories among the government researchers is that UFOs are using some form of time travel.

There needs to be a way to rule that out from AI. If you have a suggestion as to how to word that, feel free to make it.

AGI created by an un-intentionally self improving AI?