ben_w

16 ms·

271.
by ben_w 2mo ago
On the one hand, yes; on the other hand, so much of the training data comes from scraping the web that it feels wrong for them to do what they deny others the right to do. On the third hand, the settlement Anthropic famously had to pay was
272.
by ben_w 2mo ago
Interesting. I had a similar idea around a decade ago. My thoughts from the time: if it's just [something like the linked implementation], there's a limited view of the way ahead, so it's better if you can give it a non-linea
273.
by ben_w 2mo ago
They claim to have that number of MAUs. He was complaining a lot about bots when he tried to back out of the sale. He's now built a bot into it. I find myself unable to trust his statistics, not only because of his history, but also du
274.
by ben_w 2mo ago
I believe the point is the unsurprising result. That said: > Canada debanked right wing truckers Who were participating in the illegal blockade: https://www.canadianlawyermag.com/news/general/fca-upholds-2... T
275.
by ben_w 2mo ago
2025. This year's update, unfortunately, is up: https://www.carbonbrief.org/analysis-chinas-co2-climbs-2-in-...
276.
by ben_w 2mo ago
Always good when the surprise news is a win-win. Aside: You'd think the headline writer would pick a better adjective for solar anything than "Underground"
277.
by ben_w 2mo ago
You're in a world where, continuing this metaphor, 60 random Chinese companies are making and exporting home virology labs. A world where previously exported home virology labs are actively getting "upgraded" by people eager
278.
by ben_w 2mo ago
Neither was Hugging Face.
279.
by ben_w 2mo ago
It can be anything from fully automated luxury communism, to fully automated dystopian communism. From https://en.wikipedia.org/wiki/Fully_Automated_Luxury_Communi... to https://en.wikipedia.org/wiki&#x
280.
by ben_w 2mo ago
You're on the internet, your threat is everyone else running a new model with unknown capabilities without safeguards. In particular, all my last paragraph. I do offline backups, which get physically unplugged between sessions. Even
281.
by ben_w 2mo ago
Great. And the machine you used to write this comment? "your phone, laptop, etc"? Because otherwise you're not taking the threat these new models pose seriously. Catch 22, basically: anyone who thinks OpenAI should have know
282.
by ben_w 2mo ago
> Who said criminally negligent? Me. I am saying that the act of taking this standard seriously, the standard "there is no such thing as bug-free software", would classify just about every business and individual criminally neg
283.
by ben_w 2mo ago
> Do you want me to quote formal trade agreements that were made or are you able to just do a Google search yourself and see? I wrote "Those trade deals aren't worth the paper they're written on", you seem to be respo
284.
by ben_w 2mo ago
> But they are still real trade deals, the rules are still followed, &c. No, that's the point, they're not. > Your personal distaste for Donald Trump (I believe he belongs in jail and didn't vote for him either time
285.
by ben_w 2mo ago
Have you done that to your own machines? Seriously. If your reaction to the inability to know about previously unknown vulnerabilities is "unplug the Ethernet cable", why are you not doing that (and equivalent) right now to your
286.
by ben_w 2mo ago
> "it automatically has godlike powers to exploit almost any software" No it isn't. Random people on sites like this mock them as if they're talking about having godlike powers. Each new model is "merely"
287.
by ben_w 2mo ago
> We have a lot of trade deals and have created new ones while Trump has been in office. Countries have willingly negotiated with us and us with them. What you are writing is ideological wishful thinking, not what is reflected in reality
288.
by ben_w 2mo ago
> They can be repaired - but if you think they can't be, then the US should act accordingly which would mean yea go ahead and take over Greenland, withdraw forces from Europe, &c. because as you say they can't be fixed so w
289.
by ben_w 2mo ago
> Ukraine was used as a pawn to anger Russia. What did Russia try to do and continue to do so? Ukraine is a sovereign country, free to decide for themselves which military and economic bloc best suits them. Russia rolled tanks into Ukrai
290.
by ben_w 2mo ago
If this is your standard, I challenge you to name one currently operating business that isn't criminally negligent. I'm sure there's tens to hundreds of millions of them amongst the 37% of the world with no internet connect
291.
by ben_w 2mo ago
> not relied on a buggy software sandbox. Third, while we had tested and validated this sandbox, the agents were able to chain together previously unknown vulnerabilities (“0-days”) in the package management service exposed within th
292.
by ben_w 2mo ago
> Is there actually such a thing as "alignment" as a solution to that or is it just used as a name for a desired magical level of "read the mind of the entire world" that we don't know how to build and haven&#x27
293.
by ben_w 2mo ago
Hmm. I think I misunderstood what you meant by "experience policing people around financial incentives" in that case.
294.
by ben_w 2mo ago
They don't need to be omnipotent, and they're already human-or-superhuman at persuasion: https://arxiv.org/html/2411.06837v2 This may just be that humans find long arguments more persuasive than short ones, o
295.
by ben_w 2mo ago
I suspect the current models probably can find literal millions lying around for the taking, given they could pull off the incident under discussion . Tens of millions, even. Getting them to run correctly is dangling in front of the resear
296.
by ben_w 2mo ago
Cryptography is real, but nobody in that field seems to be hubristic enough to think their methods are flawless, and there's a degree of suspicion than the best models may have secret weaknesses engineered into them by the governments
297.
by ben_w 2mo ago
People have been saying that since about the invention of the internet. There's already a bunch of documented ways to exploit system hardware to jump airgaps. Bang the system bus the right way and it's a radio antenna that can dir
298.
by ben_w 2mo ago
Close; at least with a machine you can poke around inside the activations and see what it's thinking. Closest with a human is an fMRI (which is much lower resolution, though to me still bordering on the miraculous) or an implant (each
299.
by ben_w 2mo ago
Is it even a solution in the short term? It only mostly worked up until now; with models such as reported, it's felony-as-a-service if you use language a bit too hyperbolic, e.g. "we need X by the end of the day!" -> [thin
300.
by ben_w 2mo ago
> Based on OpenAI's description of the prompt, it seems to me that the computers did exactly as they were told. They were perfectly "aligned" with the stated objective and parameters of the task. The models are supposed to

More ›