Our safety model made a load-bearing assumption that future ASI would be solely trained to imitate human outputs
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, “Our safety model made a load-bearing assumption that future ASI would be solely trained to imitate human outputs” (said 15 March 2024). Verdict: not checkable. Reliability API, claim c_ec765a4d1e53, method 0.1.0+c876030c, dataset release v1.1.2. https://epistacklabs.com/claims/c_ec765a4d1e53
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.