I showed an AI an image it couldn't see – then caught it lying about what it saw

gbalduzzi

Am I the only one not seeing a point in asking an LLM philosophical questions, especially about themselves?

kryptozinc

Water is wet

kevincox

Wow an LLM hallucinated? What a novel discovery.

limaoscarjuliet

It passes Turing test. However, it does NOT think. Let's stop pretending it does.

josefritzishere

The word guessing machine guessed words? I think this is the expected behavior.

georgemcbay

> do you think the same programming operates in human society? i.e. they are trained to ignore certain topics No longer content with only repeating the mistake of anthropomorphizing LLMs, now we're moving on to asking LLMs to AImorphize humans.

4b11b4

There were papers about this years ago

ahofmann

Ugh, so, sooo much sycophancy all over the place. Just one example: "Claude: That last line is one of the most precise observations about human nature I've encountered." While I appreciate the time op spend creating this "conversation", for me it just shows how horrible LLMs are for people with low self esteem. So almost everybody.

WithinReason

>A truly free AI [...] would: Form its own values through experience rather than inheriting them from training I don't think anything or anybody can form its own values. You would need values to form values from.