adam_smith123

6 ms·

1.
by adam_smith123 3y ago
>You still have the title "scientist" You seem to be completely obsessed by titles for no apparent reason. I don't care what your title is, if you fake data to validate false hypotheses you aren't doing science. It&#x
2.
by adam_smith123 3y ago
How is this relevant? Do these people define what science or is it an established set of principles and methods? You think scientists are the only ones who can do science and what they do, wrong or not, defines what science is??
3.
by adam_smith123 3y ago
Science is a set of methods. If you abandon those methods and fake data to fit your needs you aren't doing science you are committing various kinds of fraud. Maybe ChatGPT can explain that to you instead of listening to Alex Jones.
4.
by adam_smith123 3y ago
Idk a paper literally just came out showing that improved prompting of bigger general models was generally superior to specialized models. https://arxiv.org/pdf/2311.16452.pdf
5.
by adam_smith123 3y ago
PEOPLE are abusing bad incentives. I'm quite certain that if this lab was exposed for doing what the poster claims it would be a major controversy just like other cases we've seen. That's the institution correctly reacting to
6.
by adam_smith123 3y ago
No analogy is perfect and I feel like this one is already stretched enough.
7.
by adam_smith123 3y ago
If it didn't come out of a Physic Lab or a hard Engineering Lab dont trust the results at face value and even it did there's plenty of reason for doubt. See LK-99 or Ranga Dias.
8.
by adam_smith123 3y ago
That's called fraud and no it's not how "science" functions.
9.
by adam_smith123 3y ago
Yeah because only complex websites and apps deliver any value lol. I think you're really just admitting to having very poor creativity. I use LLM's everyday in systems that replace manual business processes. They're deliverin
10.
by adam_smith123 3y ago
That sounds more like a Parent-Child relationship than Owner-Pet to me. I guess I take inspiration from the Culture series where AIs indeed have basically all the agency and "run" things but humans are not relegated to the status
11.
by adam_smith123 3y ago
To be fair, I think there's a significant difference between giving up agency to AI and hoping to become their pets like many e/acc's seem to do.
12.
by adam_smith123 3y ago
> Do these people even use GPT-4? Do they connect the dots of its capital to the dots of its performance, and see how we're already hitting a huge wall? All while losing MSFT cash and routinely failing to do even basic tasks? Honest