For typical SOTA AI systems in 2028, will it be possible for users to know the true reasons for systems making a choice?
28
Never closes
Very Unlikely (<10%)
Unlikely (10-40%)
Even odds (40-60%)
Likely (60-90%)
Very likely (>90%)
For typical state-of-the-art AI systems in 2028, do you think it will be possible for users to know the true reasons for systems making a particular choice?
By “true reasons” we mean the AI correctly explains its internal decision-making process in a way humans can understand. By “true reasons” we do not mean the decision itself is correct.
This question is managed and resolved by Manifold.
Get 
1,000 to start trading!
1,000People are also trading
Related questions
[Carlini questions] SOTA AI systems still regularly "hallucinate" incorrect solutions to problems
[Carlini questions] A highly-popular SOTA AI system is supported by advertising
What will be true of the SOTA AI on the FrontierMath benchmark, before 2026?
What will be true of the SOTA AI on the FrontierMath benchmark, before 2028?
[Carlini questions] Cost of single SOTA AI system training run by 2027
63
Will a SOTA open-sourced LLM forecasting system make major use of quasilinguistic neural reps (QNRs) before 2027?
19% chance
What will be true of the SOTA AI on the FrontierMath benchmark, before 2027?
By 2026, the SOTA in image generation will be using a voice chat to control the generation.
49% chance
[Carlini questions] Cost of single SOTA AI system training run by 2030
148
By 2026, the SOTA in image generation will be using mind reading to control the generation.
15% chance