For the entire span of my life, people have tried to develop AI systems and anticipated a day when those systems can pass the Turing Test.
Now that day has arrived, and no one seems to care about Turing Tests anymore. Why not?
Is it that
1) We're not actually there? It would take more than a few simple patches (google the answer to arithmetic questions!) on top of ChatGPT to pass a turing test?
2) Arriving there makes it clear that the Turing Test never was the right metric?
3) We're way to busy trying to figure out how LLMs are going to upend out lives to be worrying about something so academic?
4) The entire misinformation age has soured us on the idea of something like a Turing Test for ascertaining anything?
Something else?
5) Maybe we were presumptuous to suppose we're intelligent ourselves and that what we imagine to be intelligence is in its own way as shabby a trick as LLMs?
@pieist @ct_bergstrom hahaha fuck right off, this is the same trite bullshit as Sam Altman's "I am a stochastic parrot".
Wake me up when chatgpt will have produced anything as innovatively new as calculus, general relativity, cubism, microbiology... or even anything remotely new for that matter.
Please don't mistake your inability to come with nothing better than badly parroting a tech guru with the extent of what the human mind has been able to provably do for millenia
@maxx
Sorry I didn't provide a trigger warning for you. Nurse will come soon with some soup.
@maxx @pieist @ct_bergstrom I'm curious. When did YOU produce anything as innovatively new as calculus, cubism, etc.? Are you saying that anyone who fails to innovate at that level has failed the test of being truly intelligent?
@sheldonrampton @pieist @ct_bergstrom fallacy.
Some humans have, and it can be reasonably expected that some will.
No llm ever have, and nothing so far indicates that it can be expected that any ever will.
See the difference? It's enough to discard the argument. My, your, or anyone's personal ability or lack thereof is not at play here.
@maxx @pieist @ct_bergstrom No, I don't see your point. You're making arbitrary assumptions to support an an arbitrary. entirely speculatiive comparison between a handful of past exceptional human acts of genius (drawn from a population of billions of human beings) against discoveries that you CLAIM arbitrarily will never be possible from LLMs, even though LLMs are still quite new and advances in AI techology are accelerating. You can think what you want, but it's not a fact-based comparison.
@sheldonrampton @pieist @ct_bergstrom Dude 1 says : maybe humans are no more than llms.
I say : incorrect, we have numerous examples of things that human have done that llm haven't (and yes I chose undeniable feats on purpose). Humans are different and better.
You say : are you the smartest human ever? You know there are dumb humans.
To which I respond that it is irrelevant.
@maxx @pieist @ct_bergstrom The problem is that ChatGPT was literally released less than a year ago, and you're trying to compare its accomplishments during that brief period against your cherry-picked list of the best achievements of billions of humans during thousands of years. It's not a valid comparison.
@sheldonrampton @pieist @ct_bergstrom Wait wait wait… do you believe chatgpt learns anything or gets better on its own through time?
@maxx @pieist @ct_bergstrom Well, we're up to ChatGPT 4 already, and new versions are on the way. I saw a big improvement in 4 vs 3 or 3.5. The developers have said that they're using interactions with users to improve the model, and I believe them. So yes, I do believe it has gotten better over time and will continue to improve. Whether it does that "on its own" depends on what you mean by "on its own," but human beings also get a lot of assistance with their intellectual progress.
@ct_bergstrom @pieist - I’m with you on this one. That thought hit me hard recently when I saw those pictures from the James Webb Space Telescope.