hgoel

5 ms·

31.
by hgoel 1mo ago
I wonder how many innocent children America will have to murder for this case...
32.
by hgoel 1mo ago
For personal needs I use a local Qwen3.8-Next-Flash setup on a GB10 cluster. For work, Github Copilot with either GPT 5 mini or toss up between Opus/Sol depending on the complexity of the task. Used to pay for a Claude 20x plan and did
33.
by hgoel 1mo ago
So, as usual, the proposal is to limit what average people can do despite them not having behaved incorrectly nor having the capital to achieve the scaling of the big players, when the big players are the ones causing the harm? Not to menti
34.
by hgoel 1mo ago
Anthropic should become a case study in how to destroy good will in a short period of time. Ever since the spat with the DoD, they've behaved poorly almost weekly, almost making OpenAI seem better (but not really).
35.
by hgoel 1mo ago
I don't get it, you guys speak as if you have infinite time and no goals besides just the act of writing code, with no consideration about what the code you're writing is supposed to do. As if you'd be happy to just write a
36.
by hgoel 1mo ago
I'm not saying they did the hacking intentionally, I'm saying they're intentionally playing loose with the obvious safety measures to make AI seem more dangerous than it is.
37.
by hgoel 1mo ago
I wonder how much of this is intentional "incompetence" so they can justify the most recent campaign to build a regulatory moat against competition. The repeated refusals to disclose until caught certainly seem malicious, yet at t
38.
by hgoel 1mo ago
I don't think you can do much about it if you're hoping to hide that you have been in the domain for less than a year. You're going to have to admit your inexperience and ask questions of your more experienced coworkers and t
39.
by hgoel 1mo ago
Sam also said Astra was AGI and he's notoriously honest. This is just their way of saying they're officially switching from providing AI to using the AI to compete with their customers, a la Navier-Stokes drama.
40.
by hgoel 1mo ago
I also work at an intersection of tech, applied research and science. My experience has been different. It's been surprising how controlled my collaborators/coworkers have been with their reliance on AI. Sure, they use it a lot, b
41.
by hgoel 1mo ago
Licensing for selling anything with a computer in it would cripple competition from small incumbents. They're mostly just regulatory capture for the established players that created the problem in the first place. Age restrictions are
42.
by hgoel 1mo ago
I don't get how instagram is as popular as it is. I made an account last week to try to look for pictures from an event I recently attended. First I was incredibly surprised and frustrated that their search functionality is effectively
43.
by hgoel 1mo ago
Having had a similar upbringing, I strongly agree. I have a hard time not lashing out at people that fall into this trap given how important that early access to computers ended up being to achieving my dreams.
44.
by hgoel 1mo ago
I think the concern is a forced bit of functionality in all operating systems. It would quickly lead to a "and you need a license to sell your device so we can check that it can't bypass mandatory chuld-safety guards". Just l
45.
by hgoel 1mo ago
How do you reconcile this supposed caution with getting filthy rich in the process of exposing your loved ones to this risk? Did you only recently become aware of this risk? And semi-related, how do you reconcile this caution with the reckl
46.
by hgoel 1mo ago
I'm not fully convinced about the greed explanation. It seems to be unrealistic to me that greed can be at a level that the AI frontier (at least in the West) almost uniformly agrees (often with a smug smile,) that they are actively wo
47.
by hgoel 1mo ago
If we were to take the nuclear analogy, what's happening in AI right now is where the people selling nuclear power make a ton of noise about how they need the power to regulate their competitors because nuclear bombs might set the atmo
48.
by hgoel 1mo ago
The companies talking the most about safety and regulations aren't even properly taking the obvious measures. Shows that it's more of a marketing thing than something they take seriously.
49.
by hgoel 1mo ago
Yeah, the latest batch of <256GB Chinese models are really nice. They're far less cryptic than Claude, and competent enough to feel almost near Opus. I canceled all my subscriptions and switched to running the Chinese models locally
50.
by hgoel 1mo ago
As a statement of fact divorced from context, this is of course true, but it's worth putting it in context of what small-medium scale models have been achieving recently. Many of the most recent releases from Chinese labs are almost on
51.
by hgoel 1mo ago
Especially after the blatant cover up of their uncontrolled bot swarm infesting the internet, and the feckless "hopefully we do better" response upon being caught, I don't think OpenAI deserves much grace until they properly
52.
by hgoel 1mo ago
They can't exactly occupy the same space due to the Pauli exclusion principle. IIRC that's believed to be the final "barrier" that prevents neutron stars from collapsing into black holes. But this is also getting into th
53.
by hgoel 1mo ago
Hmm interesting, thanks for the link, I'll have to give that a read.
54.
by hgoel 1mo ago
I'm referring to models where the actual traces are available. Eg. I've been using a local Qwen3.8-Next-Flash lately.
55.
by hgoel 1mo ago
GCN was such a promising compute architecture, AMD even pioneered stuff like async compute and compute shader heavy rendering pipelines, only to never seriously go beyond that on consumer gear. I agree with your assessment that the story of
56.
by hgoel 1mo ago
Same here, it's one thing if I'm just screwing around, but if I'm trying to do anything serious, I need to at least have a handle on what it's doing, and, when thinking traces are available, keeping track of any logical
57.
by hgoel 1mo ago
Social media is already full of hentai art of that sort and that isn't a problem to most countries, so a little thought leads to the obvious point that your basic assumptions are incorrect. Minors were involved and people were generati
58.
by hgoel 1mo ago
Prompting for some of the results was almost the "Computer, do a breakthrough. Make no mistakes." meme. Just someone telling the model to keep trying a couple of times. Unfortunately we don't actually know what kind of prompt
59.
by hgoel 1mo ago
Yeah, I suppose we should hold off on teaching them everything that isn't critical thinking, lest they develop skills other than what is deemed acceptable by people like you. Maybe we should have them live in 15th century lifestyle tow
60.
by hgoel 1mo ago
It depends on the time the job itself takes. If you're having the LLM handle a training run for another model, the LLM is probably spending most of its time waiting for iterations rather than consuming tokens. For a task I left a local

More ›