gs17

12 ms·

301.
by gs17 1y ago
> requiring everybody else who opens their link to load the wrong page, edit it It's not required. You have a preference for the "new" experience, but many people see the additional time it takes to load and read the &quot
302.
by gs17 1y ago
> [A] An agent can process untrustworthy inputs > [B] An agent can have access to sensitive systems or private data > [C] An agent can change state or communicate externally Somewhat reminds me of the CAP theorem, where you can pic
303.
by gs17 1y ago
I suspect it's a bespoke script. The first use outputs "[domain] -> [ip]", the second use outputs "[domain] [http code] [?] [size?] [title] [info]".
304.
by gs17 1y ago
Especially now that OpenAI is entering that space officially by the end of the year.
305.
by gs17 1y ago
Worse, scanning the mask would guarantee no match to a citizen's face.
306.
by gs17 1y ago
I wouldn't worry too much about it breaking anything with how version-specific modding already is. And by the time the full release is out, I'm sure every tool will have updated based on the new names from the snapshots.
307.
by gs17 1y ago
I doubt they care that much, since the cost of a false positive is that user having to verify their age and the alternative is all users having to verify their age.
308.
by gs17 1y ago
I don't think the "use technology to detect underage users" approach has ever worked well for its stated intent (it works okay for being able to say you're "doing something" about the problem while not losing a
309.
by gs17 1y ago
I bet it's how many people trigger the "safety" filter, which is way too sensitive: https://www.reddit.com/r/ChatGPT/comments/1ocen4g/ummm_okay_...
310.
by gs17 1y ago
> What would be more interesting is how effective ChatGPT is being in guiding them towards other ideas. Most suicide prevention notices are a joke - pretending that "call this hotline" means you've done your job and that&#
311.
by gs17 1y ago
On my MacBook it's not "insanely small", but I zoom to 120% for a much better experience. I can read it just fine at the default.
312.
by gs17 1y ago
I think the easiest way to have 24/25 of something would be to have the "part 2" for each puzzle get released the next day. It would probably ruin the momentum about as much as having days off would (as another alternative to
313.
by gs17 1y ago
I miss Cardboard too. That little magnet "button" was one of the most creative ways to create an input method I've ever seen.
314.
by gs17 1y ago
The "scam" part isn't the radiators, it's the launch price. Their whole cost estimate depends on a different company reducing prices at least 50x.
315.
by gs17 1y ago
> I still believe that if Microsoft had forced Samsung to release the Odyssey VR headset worldwide, WMR could have been a success. I'm not sure if Microsoft actually wanted to try to make it a success. They made a lot of decisions
316.
by gs17 1y ago
It is, unless you take Musk's hype about Starship as fact. With rockets that are actually potentially available the best price is $1500/kg to LEO, so either they're presuming the whole setup weighs in at 3-4 tons (which is le
317.
by gs17 1y ago
They taught me that Sir Isaac Newton is the deadliest son-of-a-bitch in space. Which is probably something else these space data centers will struggle against, it'll be interesting to see how much shielding they have against impacts. T
318.
by gs17 1y ago
I had no idea how bad it had gotten, since none of it was being shown to me. Then a friend sent me his AI music on Spotify, I listened, and my recommendations were all AI for my usual genres suddenly. It's like they have a flag for you
319.
by gs17 1y ago
Especially when it's just a base64 decode directly piped into bash.
320.
by gs17 1y ago
I don't know how it's evolved since, but yeah, WPS was an almost perfect clone of Office a decade or so ago. I could help Chinese students use it because everything was so similar that the UI was usable without knowing Chinese if
321.
by gs17 1y ago
They were talking about within one model, not between AMD and Nvidia.
322.
by gs17 1y ago
In the article: > Casserly said that tasks will not be related to any of Uber’s autonomous partnerships or the development of driverless vehicles. But also it would be hard to provide them with tasks that are genuinely useful to training
323.
by gs17 1y ago
One I went to had a flock of magpies taking turns eating cat food. Different crowd during the day I guess.
324.
by gs17 1y ago
30 million years of what? Go back a few centuries and you could say the same statement about slavery, since all of civilization is a rounding error on tens of millions of years.
325.
by gs17 1y ago
Yeah, I've been keeping Perplexity around since I get a $5/mo education plan and it gives access to everyone's models (plus search) at once.
326.
by gs17 1y ago
I can explicitly tell it to search and it won't.
327.
by gs17 1y ago
Which interface are you using for it? I use the gemini.google.com one and most of the time instead of searching it at most pretends to search and hallucinates the result.
328.
by gs17 1y ago
That's my experience as well. Gemini doesn't seem interested in doing searches outside of Deep Research mode, which is kind of funny given it should have the easiest access to a top search engine.
329.
by gs17 1y ago
It's funny you took it that way, I took it as dismissive at Barbados for being smaller, and it's not clear which one was actually intended.
330.
by gs17 1y ago
> but I remember my Chinese visa being sort of a headache to deal with 10 years back from the US I got one recently and it's not bad, except that it needs to be done in-person at an embassy based on the state you live in, so there&#

More ›