xiphias2

8 ms·

91.
by xiphias2 6mo ago
It was from GPT-2 and Dario was part of the developers of that model while he was working in OpenAI, not Sam Altman, it's his playbook
92.
by xiphias2 6mo ago
It's against xAI Colossus datacenter in Memphis
93.
by xiphias2 6mo ago
How does this compare to DFlash? https://z-lab.ai/projects/dflash/ And DDTree? https://liranringel.github.io/ddtree/
94.
by xiphias2 6mo ago
AI is unstoppable as the algorithmic changes are at least as fast as compute improvements. One proof of it is that even older GPUs' value is going up.
95.
by xiphias2 6mo ago
They have a choice: study chess and beat Magnus. Until then I will care about Magnus (and lichess) more then those businesses. The best thing they did was that they bought an amazing domain name.
96.
by xiphias2 7mo ago
Isn't treesitter integrated to nvim anyways at this point, even if it's experimental support?
97.
by xiphias2 7mo ago
Because it wouldn't look good for the EU
98.
by xiphias2 7mo ago
Another project without running real benchmarks. It's very easy to generate tokens, it's much harder to solve tasks locally.
99.
by xiphias2 7mo ago
It doesn't look like RAM, CPU GPU or bandwidth is getting cheaper if that helps you, quite the opposite.
100.
by xiphias2 7mo ago
It's not about liking / disliking, but conservatives tend to prefer staying together even if it's a bad relatioship, and liberals prefer splitting by default if there are serious problems. The syncopath style is clearly categ
101.
by xiphias2 7mo ago
He was talking about it in the Lex Friedman interview after Trump was elected. And he was talking about a lot of things the Biden administration forced on Silicon Valley at that time (since then Google lost a case about one of these back-de
102.
by xiphias2 7mo ago
Marc Andereseen has talked about the downside of RLHF: it's a specific group of liberal low income people in California who did the rating, so AI has been leaning their culture. I think OpenAI tried to diversify at least the location o
103.
by xiphias2 7mo ago
It's not really shameful, if you want LIDAR, just buy a new Xiaomi SU7. What's much worse is that Tesla is hiding the reasons for the crashes in the crash reports (the only company with FSD).
104.
by xiphias2 7mo ago
Probably it's true from reputation standpoint. Some people in Europe were not that happy when Biden told on public television that the Nord Stream pipeline will be blown up somehow, but luckily the media was good in not talking too muc
105.
by xiphias2 7mo ago
> The right way might be to fight AI slop with AI enforced guard rails. Whenever I you tried to develop using guardrails with LLMs, I found out that they are much better at ,,cheating'' than a human: getting around the guardrai
106.
by xiphias2 7mo ago
There are even simpler things: the rating system. There's no guarantee that the driver won't see what I rated him, so I won't report them. There are ways to report if a man has been sexual with a woman, but they somehow just
107.
by xiphias2 7mo ago
The main problem I see is adding things slowly instead of automatic rewrites. I remember adding lifetimes in some structs and then wanted to use generics and self pointing with lifetimes because that made sense, and then it didn't work
108.
by xiphias2 8mo ago
It's a product category name at least, not a release name, so the next release can be Neo 2
109.
by xiphias2 8mo ago
This is just not true, people get promoted for delivering impact whether the solution is complex or simple. The best engineer I know who can work with huge complex systems in a big company usually starts with a complex solution then after h
110.
by xiphias2 8mo ago
The main lesson is quite simple: if you can write the test to be uncheatable, ChatGPT can write the code for it.
111.
by xiphias2 8mo ago
Is there a way to get sticky model selection back, or the reason is that it is just too expensive to serve alternative models? For coding I love codex-5.3-xhigh, but for non-coding prompts I still far prefer o3 even if it's considered
112.
by xiphias2 8mo ago
Sure, and actually the open models are already good enough to do that, it's not like any company could stop any organization that can collect the data from doing this. They can just improve on it a lot.
113.
by xiphias2 8mo ago
I don't really understand this reasoning actually: if OpenClaw usage go up, and a service (OpenAI it looks like) gets lots of usage data for personal assistent usage, they can optimize to make it better for people who get a $200 subscr
114.
by xiphias2 8mo ago
Probably drones targeting and automatically killing Russian people by a thinking model guessing if its Russian on Ukrainian person is a red line. Elon Musk already denied Starlink for being used for remote killing, but at some point all the
115.
by xiphias2 8mo ago
I agree completely. Anybody who uses the models extensively know it can do something amazing for a prompt and something awful for another. But I also know that wars are unfortunately real and there are real enemies between countries and the
116.
by xiphias2 8mo ago
For anybody who thinks it's about Trump vs other administration: it's not, both AI surveillance of all people and using it for automatic fight was just bound to happen. The only question is whether the safety of the models were r
117.
by xiphias2 8mo ago
I understand that there's a precedent here, but isn't normally the precedent for the opposite in contract law? And if UK is precedent based, how come the previous precedents don't apply here? I agree that no toilet breaks is
118.
by xiphias2 8mo ago
Congrats! Great try! I have a different view point on what to automate and I'm working differently with agents, but I much prefer seeing projects like this on HN to just product announcements.
119.
Show HN: LazyViewer – TUI code viewer with Git diff previews(github.com)
3 points by xiphias2 8mo ago|0 comments
120.
by xiphias2 8mo ago
,,Needless to say, I support Anthropic here. I’m a sensible moderate on the killbot issue (we’ll probably get them eventually, and I doubt they’ll make things much worse compared to AI “only” having unfettered access to every Internet-enabl

More ›