Comment by stickfigure
5 hours ago
I think it's too early to declare the Turing test passed. You just need to have a conversation long enough to exhaust the context window. Less than that, since response quality degrades long before you hit hard window limits. Even with compaction.
Neuroplasticity is hard to simulate in a few hundred thousand tokens.
"You're absolutely right!"
I think for a while the test was passed. Then we learned the hallmark characteristics of these models, and now most of us can easily differentiate. That said -- these models are programmed specifically to be more helpful, more articulate, more friendly, and more verbose than people, so that may not be a fair expectation. Even so, I think if you took all of that away, you'd be able to differentiate the two, it just might take longer.
Right. I think the modern LLMs are quite good at mimicking human words, but we were initially taken in like we were in the 1960s by ELIZA. It’s a (increasingly sophisticated) magic trick, but it’s just a trick.
There are a lot of difference kinds of LLMs. 0 of the ones I've encountered are good writers, in fact all of them are horrible at it.
But I wonder if there's one out there that I don't know about with a different kind of training that actually is good at writing and fun to talk to for a long time. (granted somepeople love talking to gpt 4, but also some people loved talking to ELIZA so clearly some people have a super high tolerance for slop.)
It's weird, I don't know how normally pedantic comp sci. people let this meme that the Turing test is beaten by LLMs to spread so unchallenged. As far as I'm aware, there is no restriction in the Turing test that demands that the interrogator be ignorant of the latest state-of-art in computing (and AI tech), nor is there a strict time limit enforced for the questioning?
Given these conditions, it should be relatively easy for the interrogator to expose the AI in this current day and age.
It was not meant as a pass/fail
For as rigorous of a Turing test as you present, I believe many (or even most) humans would also fail it.
How many humans seriously have the attention span to have a million "token" conversation with someone else and get every detail perfect without misremembering a single thing?
Response quality degrades long before you hit a million tokens.
But sure, let's say it doesn't. If you interact with someone day after day, you'll eventually hit a million tokens. Add some audio or images and you will exhaust the context much much faster.
However, I'll grant you that Turing's original imitation game (text only, human typist, five minutes) is probably pretty close, and that's impressive enough to call intelligence (of a sort). Though modern LLMs tend to manifest obvious dead giveaways like "you're absolutely right!"
But context window exhaustion does not look like mere forgetfulness, but more like loss of general coherence, like getting drunk.
Doesn't the Turing test require a human too, to be compared to the AI?
I don't know. Practically, LLMs are already better conversation partners on any topic compared to the average human I have access to. This also holds in reverse, of course - if someone wants me to explain something, usually they'd be better off asking an LLM.