ClaimsEliezer Yudkowsky

“the pretty picture of the girl with the with the steampunk goggles on her head if I'm remembering correctly”

Eliezer Yudkowsky, interviewed by Lex Fridman March 2023 "Dangers of AI and the End of Human Civilization" Yudkowsky on Lex Fridman (LessWrong) archived copy

Factual AI capabilities

The image generated from Bing Sydney's self-description was a pretty picture of a girl with steampunk goggles on her head.

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

True: the widely shared image generated from Bing/Sydney's self-description (early 2023) shows a girl or young woman with steampunk-style goggles on her head. Mostly true: it shows a young woman with goggles on her head, but a minor detail is off (e.g. the goggles are not clearly steampunk, or they are worn slightly differently). Mixed: the image matches only some elements, e.g. a girl with no goggles, or goggles on a non-human figure. Mostly false: there was such an image, but it differs substantially (e.g. no girl and no goggles on the head). False: no image generated from Sydney's self-description resembled this, or the circulated image is clearly something else.

Set before the research began.

Evidence (3)

  1. E1 Bing Chat imagines itself as a human girl and we get a face to put to Sydney Windows Central · 11 April 2023

    “Bing Chat put a face to itself and showed Reddit user SnooDonkeys5480 what it imagines it would look like as a human girl.”

  2. E2 Bing / Sydney (Prometheus) — Pantheon Pantheon (llm-pantheon.org)

    “2023-02-18 @repligate — “A few of Sydney’s self portraits.””

  3. E3 AI #3 Don't Worry About the Vase (Zvi Mowshowitz, WordPress) · 9 March 2023

    “as is The Once and Face of Sydney.”

Reasoning

E1 reports that Bing Chat showed what it imagined it would look like as a human girl. However, it does not mention goggles, and it is dated April 2023, after the claim was made, so it may refer to a different image. E2 lists a February 2023 post of Sydney's self-portraits but does not say what they show. E3 mentions "The Once and Face of Sydney" without describing the image. None of the evidence confirms or rules out the steampunk-goggles detail.

Cite this claim

Eliezer Yudkowsky, “The image generated from Bing Sydney's self-description was a pretty picture of a girl with steampunk goggles on her head.” (said March 2023). Verdict: unresolvable. Reliability API, claim c_e6f3bb04559d, method 0.1.0+c876030c, dataset release v1.1.2. https://epistacklabs.com/claims/c_e6f3bb04559d

This claim is in dataset release v1.1.2, with the same verdict.

Reply to this claim

For Eliezer Yudkowsky, or someone speaking for them: a correction or counter-evidence. Replies are read before anything is shown; an approved reply shows your name, role, date, statement and link. Your email address is never shown and is deleted 30 days after the decision (privacy).

More factual claims by Eliezer Yudkowsky

  1. Resolved true Factual September 2025
    “We learned that that's false, but we did not learn that you can have a thing that can redesign a tree and build an AI much smarter than itself, that builds an AI much smarter than itself, and that…”

    No AI has yet been shown to redesign a tree and build a much smarter AI that builds a still smarter AI; that scenario has not occurred.

  2. Mostly true Factual 25 March 2024
    “I was a passionate noticer of the human brain not being homogenous, and was therefore unpleasantly surprised by transformers scaling as well as they did.”

    Yudkowsky was unpleasantly surprised by how well transformers scaled, having believed the human brain is not homogeneous.

  3. Mostly true Factual 6 April 2023
    “First of all, you realize in principle that the task of minimizing losses on predicting human text does not stop when you’re as smart as a human, right?”

    In principle, the task of minimizing loss on predicting human text does not stop being improvable once a system is as smart as a human.

  4. Mostly true Factual 30 March 2023
    “Transformers are the main thing like that.”

    Transformers were the main architectural change that gave a non-linear jump in performance, compared with earlier approaches such as RNNs.

  5. Mostly true Factual 30 March 2023
    “Because the basic paradigm of machine learning that we are presently operating under is that you can have the loss function, but only for things you can evaluate.”

    The basic paradigm of machine learning currently in use is that a loss function can be defined only for things that can be evaluated.

All of them

Verdicts are made automatically; how verdicts are made.