This is like literally my AI alignment fantasy from 2003, though not with RLHF as the implementation method or LLMs as the base.
Not checkable An opinion, value judgement or statement with nothing to check.
Not checkable: An opinion, value judgement or statement with nothing to check. Not scored.
Cite this claim
Eliezer Yudkowsky, “This is like literally my AI alignment fantasy from 2003, though not with RLHF as the implementation method or LLMs as the base.” (said 6 April 2023). Verdict: not checkable. Reliability API, claim c_274380a2708b, method 0.1.0+c876030c, dataset release v1.1.2. https://epistacklabs.com/claims/c_274380a2708b
This claim is in dataset release v1.1.2, with the same verdict.
Reply to this claim
Verdicts are made automatically; how verdicts are made.