Yudkowsky said it was his "pretty strong guess" that secret government AI labs, which he said probably exist, lack the best people and could not have created ChatGPT on their own, adding "This is just a guess."
Unresolvable The public record could not settle it.
Factual: Statements about the present or the past, checked against the record: did they hold up?
What would count as true or false
The proposition is only about what Yudkowsky said; whether government labs could build ChatGPT is not being tested. True: a transcript or recording from about February 2023 shows him saying secret government AI labs probably exist, that his 'pretty strong guess' is they lack the best people and could not create ChatGPT on their own, and adding 'This is just a guess' or an equivalent. Mostly true: he said this in substance, but a secondary element (e.g. the exact 'just a guess' wording or the remark that such labs probably exist) is worded differently or missing. Mixed: only part of the attributed statement is in the source (e.g. the guess about ChatGPT but not the remark about the labs existing, plus other changes). Mostly false: he discussed government AI labs but said something materially different (e.g. no claim about ChatGPT capability). False: the source shows he said the opposite (e.g. that such labs could build ChatGPT), or the quote is not his.
Set before the research began.
Evidence (1)
-
E1 Yudkowsky on AGI risk on the Bankless podcast EA Forum · 13 March 2023 Own record
“This is just a guess.”
Reasoning
The only evidence, E2 (an EA Forum transcript of the Bankless podcast), quotes just the line "This is just a guess." The quote itself does not include the core statement: that secret government AI labs probably exist, lack the best people and could not create ChatGPT on their own. The note under the quote says the line follows a ChatGPT remark, but the note is only a pointer and cannot stand in for the quote. Because the main part of the attributed statement is not in the quoted evidence, the claim cannot be confirmed or refuted.
Cite this claim
Eliezer Yudkowsky, “Yudkowsky said it was his "pretty strong guess" that secret government AI labs, which he said probably exist, lack the best people and could not have created ChatGPT on their own, adding "This is just a guess."” (said February 2023). Verdict: unresolvable. Reliability API, claim c_498e4f8f7659, method 0.1.0+c876030c, dataset release v1.1.2. https://epistacklabs.com/claims/c_498e4f8f7659
This claim is in dataset release v1.1.2, with the same verdict.
Reply to this claim
More factual claims by Eliezer Yudkowsky
-
Mixed
“I was observing that obvious-seeming hypothesis myself, sometime around the time of GPT 3.5.”
Yudkowsky himself observed the persona-selection-like hypothesis around the time of GPT 3.5.
-
Mostly true
“The test-passing preference stands on its own once formed.”
-
Mostly true
“LLMs surely have neural circuits they learned to model text streams from fictional and nonfictional personas that are not lying, deceiving, cheating.”
-
Mostly true
“The predict-as-if-honest circuitry is contextual, invoked to predict some streams of text but not others.”
-
Mostly true
“Presumably, somewhere along the way LLMs learn a dispreference for tests that don't pass, independently of how that gets prevented or whether the new plan does what the user really wanted.”
Verdicts are made automatically; how verdicts are made.