Mind-reading AI can translate brainwaves into written text: Using only a sensor-filled helmet combined with artificial intelligence, a team of scientists has announced they can turn a person’s thou…::A system that records the brain’s electrical activity through the scalp can turn thoughts into words with help from a large language model – but the results are far from perfect
Of course they could, in the same way that hitting the autocomplete key can finish a half-completed sentence you’ve never written before.
The fact that models can produce useful outputs from novel inputs is the whole reason why we build models. Your argument is functionally equivalent to the claim that wind tunnels are intelligent because they can characterise the aerodynamics of both old and new kinds of planes.
For the same reason that a random number generator is capable of producing never-before-seen strings of digits. LLM inference engines have a property called “temperature” that governs how much randomness is injected into their responses:
Auto complete is not a lossy encoding of a database either, it’s a product of a dataset, just like you are a product of your experiences, but it is not wholly representative of that dataset.
A wind tunnel is not intelligent because it doesn’t answer questions or process knowledge/data it just creates data. A wind tunnel will not answer the question “is this aerodynamic” but you can observe a wind tunnel and use your intelligence to process that and answer the question.
Temperature and randomness don’t explain hallucinations, they are a product of inference. If you turned the temperature down to 0 and asked it the question " what happened in the great Christmas fire of 1934" it will give it’s best guess of what happened then even though that question is not in it’s dataset and it can’t look up the answer. The temperature would just mean that between runs it would consistently give the same story, the one that is most statistically probable, as opposed to another one that may be less probable but was pushed up due to randomness. Hallucinations are a product of inference, of taking something at face value then trying to explain it. People will do this too, if you tell someone a lie confidently then ask them about it they will use there intelligence to rationalize a story about what happened.
If LLMs don’t encode their training data, then why are they proving susceptible to data exfiltration techniques where they output the content of their training dataset verbatim? https://m.youtube.com/watch?v=L_1plTXF-FE
I’m not saying it doesn’t encode some of its training data, I’m saying it’s not just encoding its training data. It probably does “memorize” a bunch of trivial facts from its training data and regurgitate them when asked. I’m saying that’s not all they are and that’s not what makes the intelligent, their ability to also answer questions outside their training data is.
But they don’t “answer questions”, they just respond to prompts. You can’t use them to learn anything without checking their responses against authoritative sources you should have used in the first place.
There’s no intelligence there, just a plagirism laundromat and some rules for formatting text like a 7th grader.
It can answer questions as well as any person. Just because you may need to check with another source doesn’t mean it didn’t answer the question it just means you can’t fully trust it. If I ask someone who’s the fourth u.s. president and they say Jefferson they still answered the question, they just answered it wrong. You also don’t have to check with another source in the same way you do with asking a person a question, if it sounds right. If that person answered Madison and I faintly recall it and think it sounds right I will probably not check their answer and take it as fact.
For example I asked chatgpt for a chocolate chip cookie recipe once. I make cookies pretty often so would know if the recipe seemed off but the one it provided seemed good, I followed it and made some pretty good cookies. It answered the question correctly as shown by the cookies. You could argue it plagiarized but while the ingredients and steps were pretty close to some I found later none were a perfect match which is about as good as you can get with recipes which tend to converge in the same thing. The only real difference between most of them is the dumb story they give at the beginning which thankfully chatgpt doesn’t do.
The 7th grader and plagiarism comment make me think you haven’t played with them much or really tested them. I have had it write contracts, one of which I had reviewed by a lawyer who only had some small comments, as well as other letters and documents I needed for my mortgage and buying a home. All of these were looked over by proffesionals and none of them realized it was a bot. None of them were plagiarized too because the parameters I gave it and the output it created were way too unique to be in its training set.
Of course I have, my employer has me shoehorning ChatGPT into everything, and I agree with what the research says: Children can answer questions better than LLMs can.
https://techxplore.com/news/2023-12-artificial-intelligence-excel-imitation.html
Stochastic plagirism is still plagirism.
That study is like giving a written test to an illiterate adult, seeing them do worse than a child and saying they aren’t intelligent or innovative. Like I said earlier intelligence is multi-faceted, and chatgpt excels at rhetorical, conversational and other types of written intelligence. It does not, as that study shows, do well in spatial manipulation, that doesn’t mean it’s not intelligent. If you gave that same test to a paralyzed blind person with little to no concept of spatial reality they’d probably do just as bad. If you asked them to compose a short story or an essay they might be good at it because that’s where they’re capabilities lye. That short story could still be innovative in its composition and characters, and could be way better than anything a child wrote.
You have to measure different types of intelligence with different tests. If you asked chatgpt and a set of adults and children to write a short story about a wholey new subject chatgpt would beat most of the children and probably some of the adults.
And if that short story is about a new subject matter completey out of its training set what/who is it plagiarizing from? You could say it’s taking common tropes, themes and story elements from other stories, but that’s fundamentally what a lot of writing and culture is. If that’s plagiarism then you should be more worried about the marvel franchise as it’s a plagiarism machine that has way more cultural impact.