In 1 years time, what credence will John assign to the field of alignment converging toward primarlity working on decoding the internal language of neural nets? | Manifold

In 1 years time, what credence will John assign to the field of alignment converging toward primarlity working on decoding the internal language of neural nets?

6

110Ṁ99

Jun 1

28%

chance

1D

1W

1M

ALL

https://www.lesswrong.com/posts/BzYmJYECAc3xyCTt6/the-plan-2022-update
Resolution will be determined by the earliest statement John makes publically on his credences after 1st January 2024.

If, by 1st June 2024, John doesn't make any such statement, this question will result as N/A.

️ AI Alignment

Get

1,000

to start trading!

Sort by:

"Put all that together, extrapolate, and my 40% confidence guess is that over the next 1-2 years the field of alignment will converge toward primarily working on decoding the internal language of neural nets."

Absent other information, I will take this estimate at face value.

@Conflux Makes sense

Related questions

Will we solve AI alignment by 2026?

Will I think that alignment is no longer "preparadigmatic" by the start of 2026?

Will tailcalled think that the Brain-Like AGI alignment research program has achieved something important by October 20th, 2026?

Will some piece of AI capabilities research done in 2023 or after be net-positive for AI alignment research?

Conditional on their being no AI takeoff before 2030, will the majority of AI researchers believe that AI alignment is solved?

Will >= 1 alignment researcher/paper cite "maximum diffusion reinforcement learning" as alignment-relevant in 2025?

In 5 years will I think the org Conjecture was net good for alignment?

Will we find polysemanticity via superposition in neurons in the brain before 2040?

Will OpenAI announce a major breakthrough in AI alignment in 2024?

Will Inner or Outer AI alignment be considered "mostly solved" first?

Related questions

Will we solve AI alignment by 2026?

Will >= 1 alignment researcher/paper cite "maximum diffusion reinforcement learning" as alignment-relevant in 2025?

Will I think that alignment is no longer "preparadigmatic" by the start of 2026?

In 5 years will I think the org Conjecture was net good for alignment?

Will tailcalled think that the Brain-Like AGI alignment research program has achieved something important by October 20th, 2026?

Will we find polysemanticity via superposition in neurons in the brain before 2040?

Will some piece of AI capabilities research done in 2023 or after be net-positive for AI alignment research?

Will OpenAI announce a major breakthrough in AI alignment in 2024?

Conditional on their being no AI takeoff before 2030, will the majority of AI researchers believe that AI alignment is solved?

Will Inner or Outer AI alignment be considered "mostly solved" first?

© Manifold Markets, Inc.•Terms + Mana-only Terms•Privacy•Rules