> Teachers make mistakes, yes. But can you swap a teacher with another teacher without a student knowing? No, you can't. You can with a model and that is dangerous. And not only the company. If Ello wants to they can move students to a Chinese model and nobody would know the difference. Or the OAI model they've chosen can be dilluted or changed without Ello even knowing.
Sounds like this issue could be solved by an open-source teaching harness.
> But the ground reality is that they don't lack tutors or educators.
As someone from a developing country: you are wrong. Lack of teachers (and especially GOOD teachers) is a big issue. Lack of infrastructure is also an issue, though.
I use them every day for SWE. I cannot remember any hallucination from the last 6 months. It pretty much stopped being an issue from the beginning of this year, if you use top paid models. Free models or lower-tier paid models can still hallucinate, though.
> It's a classic example of the kinds of questions LLMs get wrong. There's plenty of others. Not sure your point here. We can easily find things a 4-9 year old will talk about that an LLM will get wrong, hallucinate, etc.
My point is that the hallucination rate of modern LLMs, while not zero, is so low that this is no longer an issue in practice.
> There's a lot of work being done in 'could'. And it's entirely ignoring the dangers.
I agree that there are some risks. But for many children, the alternative is not human teacher, but no teacher at all. This considerably changes the risk-benefit calculus, IMHO.
> I'm a mom who actually has kids and this thread is insane. 'Just get a tutor'...okay?? Are you paying for it? Because that's not an option for a lot of families. I get that it's more ideal, but the alternative is...nothing?
Well said. I think many posters in this thread come from a pretty privileged backgrounds..
> If you could time-travel back to your 5y old self, would you prefer to be taught by AI tutor given the current state o/t art, or taught by whatever teachers you did have when you were 5? (with all the existing hallucination, breaking through guardrails etc problems of current AI in mind)
I had to choose between my elementary school teachers and something like Claude Fable 5 with a good teaching-focused harness, I would definitely choose Fable 5.
> We should be wary of all new technology until it’s proven to be a benefit to society.
Nope. Any nation that thinks like this will be outcompeted by more tech-positive nations in the long-term. It's on the luddites to demonstrate evidence of harm, if they want some use of technology banned.
> Teachers don't make confident mistakes because the student asked the wrong way. Teachers can be held accountable. Teachers can learn. Teachers can love their students.
All of these can be true, and at the same it can be true that a child tutored by a good quality state-of-the-art LLM with a good teaching-focused harness could have better learning outcomes than a child without it. Even if we agreed that a good human teacher is better than a LLM, human teacher's time and attention is limited.
It is a pretty unlikely proposition that AI progress will be a net negative, especially over the long term. There is no great technology in the history of mankind that was a net negative. Even for the more controversial ones, like internal combustion engines or nuclear power, the pros still far outweight the cons. Why should AI be different?
> Intelligence is the complete opposite of an LLM. Usually the more you needed to memorize to do something the less intelligent you were considered.
Contrary to popular belief, training a LLM is not just about memorization (overfitting). There is some memorization happening, but well-trained LLMs also generalize.
> I'd rather a purpose build machine learning system that is actually designed to run them
If generalist robotic models get good enough to accomplish many varied tasks effectively, training a separate comparable specialized system from scratch for every task would be highly cost-ineffective, even if, in theory, it could have slightly higher reliability.
Yes, I do this all the time in Cline. It supports automatic model change when switching from Plan mode to Act (implementation) mode. Opus for planning and Sonnet for implementation. It works great.
Correlation does not necessarily mean causation.