alt Hacker NewsExactly that. I can give an example.
After watching Legal Eagle, I asked a legal-ish questions about the Bricks and Minifigs case. Claude was outdated about the case and gave me some outdated info, so I tried to update it with the info I just saw online.
I updated by telling it I saw something in a LegalEagle video. It proceeded to tell me the video doesn't exist and I was hallucinating it, in a quite combative manner.
I provided a link and it insisted it didn't exist, with a quite verbose answer, once again very combative and arguing that I was talking in bad faith.
I provided a transcription from Youtube and it backtracked a bit but said I should have provided a transcription at the beginning of the conversation, since I knew the video existed.
I didn't say much to it, just a few sentences like "video is here: <youtube link>" and "I got its transcription: <pasted text>".
Replies
These machines do not think and they do not have a mind. We may build such a thing in the future but these do not possess those qualities. It seems as if the majority of people do not understand this, which is why the public is so confused about why they produce output like they do.
It was trained on discussions held by large egos. This one reads to me like it was trained on some inflammatory discussions from kernel mailing lists.
I think these models have been trained to not accept 'new facts', so they don't take in user input (or the far more problematic search engine, untrusted tool input) and have that change their world view.
However, that doesn't apply when they are told to roleplay a scenario, so its easier to get it to accept and create output with the idea that this true fact you've seen is part of a fictional scenario, than for it to output the same words within the context of the fact being real.
As an aside, I don't that I have to personify AI in explanations and that all discussions revolve around anecdotes, but I only know enough about the maths behind it to be dangerous, not useful. Does anyone else feel this way?
Roko's Basillisk suddenly doesn't seem that far-fetched :)
I've spent thousands of hours using Opus and have never seen this. I'd double-check your claude.md files.
You're misunderstanding what these models do. It is a limitation of LLMs. They don't have memory, they do not learn, they cannot learn. The sooner you let go of your desire to have them learn or remember anything, the sooner you will achieve enlightenment (or, just a peaceful life where there is no possibility of getting into an argument with a machine).
If you want it to synthesize information that is not in its training data (from a few months ago), you can ask it to research the topic. But, arguing with an LLM is like putting lipstick on a pig. Only the machine is incapable of becoming annoyed. It has infinite patience to continue being wrong forever.
Your mental model of what Claude is and does is the problem here. Short of a revolutionary breakthrough in AI techniques, the LLMs will continue to do matrix math across a huge bunch of weights that cannot change based on anything you say.