olalonde

6 ms·

1.
by olalonde 3d ago
Those examples aren't really comparable in scale, especially regarding risk to consumer trust. How many users did they actually lose from wrongful death lawsuits? I guess very few. Shipping an intentionally fake privacy toggle is funda
2.
by olalonde 3d ago
Interesting. You've made several comments stating that models reason, yet here you argue they lack intelligence.
3.
by olalonde 4d ago
A bit of an aside: do you still stand by your 2022 comment that LLMs are fundamentally just a fancy search engine, or has your view changed since then? https://news.ycombinator.com/item?id=32042689
4.
OpenAI built a text generator so good, it's considered too dangerous (2019)(techcrunch.com)
10 points by olalonde 4d ago|1 comments
5.
by olalonde 4d ago
Yep: https://slate.com/technology/2019/02/openai-gpt2-text-genera...
6.
TuragaLab/flybody: MuJoCo fruit fly body model and locomotion RL tasks(github.com)
1 points by olalonde 4d ago|0 comments
7.
by olalonde 4d ago
Prediction: in a few years, we will laugh at this hysteria and call it a moral panic.
8.
by olalonde 5d ago
Not that this predates the Hugging Face hack by two months. So it's likely that OpenAI didn't know about it.
9.
by olalonde 5d ago
Of course they want the government to mandate a pause. Cutthroat competition is exhausting, and a legally enforced cartel lets everyone lock in their current market share without the pesky inconvenience of competing.
10.
by olalonde 6d ago
From OpenAI's point of view, the risks are not comparable at all: 1) Extreme edge case affecting a handful of users, vs millions of users using the data sharing opt out. 2) The deaths are unintentional. 3) They probably won't lose
11.
by olalonde 6d ago
By "loss of trust", I meant that this is something they would risk losing a lot of users over, which isn't the case with the other lawsuits. There is a massive distinction between fighting third parties in a legal grey area a
12.
by olalonde 6d ago
These examples aren't really similar. None of those situations involve harming and lying to their own customers.
13.
by olalonde 6d ago
Yes that's what I was referring to. I'm a non paid user but I did briefly pay for a consumer subscription before. Not sure what region I'm assigned to as I'm in China and always access through a VPN.
14.
by olalonde 6d ago
Weird, I'm the opposite. I don't recall ever setting mine and I just checked and it was set to "disallow training".
15.
by olalonde 7d ago
Example?
16.
by olalonde 7d ago
Unethical doesn't mean irrational. They'd be risking massive lawsuits and a total loss of trust if they got caught lying about this. Doesn't seem worth it.
17.
Ex-Anthropic researcher Jacob Coxon says AI development poses risk to humans(apnews.com)
1 points by olalonde 7d ago|0 comments
18.
by olalonde 7d ago
Container or full blown VM?
19.
by olalonde 7d ago
Are you guys all running agents in VMs?
20.
by olalonde 7d ago
Looks quite similar to the Huawei Pura X Max[0]. Funny how the cycle inverted. [0] https://consumer.huawei.com/cn/phones/pura-x-max/
21.
Larger Pacific striped octopus(en.wikipedia.org)
136 points by olalonde 8d ago|74 comments
22.
by olalonde 8d ago
Can't mathematicians still gain novel insights by reverse-engineering AI-generated proofs? Just like chess players learn new concepts by studying what engines play.
23.
by olalonde 8d ago
If their solutions are significantly different, as OpenAI claims, would it still be considered plagiarism?
24.
by olalonde 8d ago
> I wish it were done by different folks, ideally under some kind of public control like NASA research or the NPR model. This is sort of what OpenAI was supposed to be. I'll never understand how it was legal for them to turn it into
25.
by olalonde 8d ago
Isn't it safe to say that all famous unsolved math problems will get a "massive amount of AI-powered effort" pointed at them regardless?
26.
by olalonde 8d ago
OpenAI claims the data contamination issue only surfaced after they proactively reached out to Buckmaster and Alpöge to coordinate a joint release. They also say that even if there was some contamination, the underlying proofs diverge subst
27.
by olalonde 8d ago
> The agents arrived at their resolution on Saturday, September 5, about 88 hours after the first agents were launched. If this actually holds up, solving a Millennium Prize problem in 88 hours is mind-boggling.
28.
by olalonde 8d ago
Playing the devil's advocate here but it's true that OpenAI didn't have to make those offers.
29.
OpenAI Just Claimed a Math Discovery. Some Academics Are Crying Foul(wired.com)
7 points by olalonde 8d ago|2 comments
30.
by olalonde 9d ago
The scam was caught due to a whistleblower, not because people were looking for flaws.

More ›