Didn't Anthropic show that LLMs frequently hallucinate their "reasoning" steps? > Bullshitting (Unfaithful): The model gives the wrong answer. The computation

Didn't Anthropic show that LLMs frequently hallucinate their "reasoning" steps? > Bullshitting (Unfaithful): The model gives the wrong answer. The computation we can see looks like it’s just guessing the answer, despite the chain of thought suggesting it’s computed it using a calculator. https://transformer-circuits.pub/2025/attribution-graphs/biology.html#dives-cot https://transformer-circuits.pub/2025/attribution-graphs/bio...