LLMs are not like you and me—and never will be.
We recorded 10 claims from this post. 5 were checked against evidence. 5 can’t be checked. 9 more are held. Read the post at its source.
Claims from this post (10)
In the order they appear in the post. Each quotation is the speaker’s words in this post. A label such as “Supported” or “Can’t be checked” is the claim’s current assessment; its page has the evidence and the reasons.
Assessment: Can’t be checkedSaid
Observation
“That rumor is bullshit.”
Our reading: The rumor that LLMs are basically just like humans is false.
No horizon
Assessment: UnresolvedSaid
Causal claim
“Critically, as I argued at the end of June (and going back to 2019) LLMs never induce proper world models, which is why, for example, they still can’t even play chess reliably,1 and continue to make stupid, head-scratching errors with startling regularity.”
Our reading: LLMs never induce proper world models, which causes them to be unable to play chess reliably and to make frequent stupid errors.
No horizon Evidence as of
Limitation: The strongest intervention evidence concerns a specialist chess model.
Assessment: UnresolvedSaid
Observation
“They don’t induce abstract models of the world that they can reliably engage, even when those models are given to them on a silver platter (as with the rules of chess).”
Our reading: LLMs do not induce abstract world models they can reliably use, even when given the rules explicitly.
No horizon Evidence as of
Limitation: RuleArena tests practical regulations rather than chess.
Assessment: ContradictedSaid
Observation
“They do not sanity check their own work.”
Our reading: LLMs do not sanity check their own work.
No horizon Evidence as of
Limitation: SCoRe used specialized training; its results do not establish that every model checks every answer [e1].
Assessment: UnresolvedSaid
Observation
“Even with thinking mode activated, the output was filled with errors that could easily have been avoided, with a 5-second Wikipedia search that any sensible human could have performed.”
Our reading: GPT-5 in thinking mode produced a list of Canadian Prime Ministers full of easily avoidable errors.
No horizon Evidence as of
Limitation: The retrieved plain-text version of the post does not supply independently verifiable image provenance.
Assessment: UnresolvedSaid
Causal claim
“ChatGPT gets them wrong because it is munging words together, not genuinely processing an (elementary) model of British PM history.”
Our reading: ChatGPT gets the British prime minister tasks wrong because it is munging words together rather than genuinely processing an elementary model of British PM history.
No horizon Evidence as of
Limitation: Evidence from Llama-2 cannot directly support or contradict a claim about GPT-5's internal processing.
Assessment: Can’t be checkedSaid
Observation
“We still feel that way. And the field still hasn’t put the time in.”
Our reading: The AI field still has not put the time in on giving time, space, and causality a solid computational footing.
No horizon
Assessment: Can’t be checkedSaid
Observation
“But no matter how much LLMs mimic the patterns of human language, they are not like us. They sound like us, but they don’t think like us.”
Our reading: LLMs do not think like humans, however well they mimic human language.
No horizon
Qualifier: no matter how much they mimic human language
Assessment: Can’t be checkedSaid
Recommendation or value judgment
“Always regard them like the weird function approximators they are; never trust them.”
Our reading: People should never trust LLMs and should treat them as weird function approximators.
No horizon
Assessment: Can’t be checkedSaid
Observation
“The LLM itself still isn’t inducing the rules. And this is a very special-purpose hack, not a remotely general solution. (And, e.g., it would likely fall apart if you changed the rules to play an alternative version of chess.)”
Our reading: A big company's added python code for chess legality is a special-purpose hack, the LLM still doesn't induce the rules, and it would likely fail on chess variants.
No horizon
Held from this post (9)
These records did not pass every check, so nothing about them is published. Held is not a finding.
- The two reviews read the claim differently
- The two reviews read the claim differently
- The criteria are ambiguous. Our reading, not assessed: An AI system that cannot induce a robust model of chess rules will not induce what it needs to know about time, finance, and the changing world from statistical analysis of big data.
- The two reviews read the claim differently
- The two reviews read the claim differently
- The two reviews read the claim differently
- The two reviews read the claim differently
- The two reviews reached different assessments. Our reading, not assessed: Any system that could properly leverage widely available data into a proper world model could fact-check itself and would not make this error (showing two presidents in office at once), whereas LLMs leave everything to statistical chance.
- The criteria are ambiguous. Our reading, not assessed: Autocomplete, which LLMs essentially are, works well for some purposes but is not good enough for world model induction.