WhitneyLand

5 ms·

31.
by WhitneyLand 3mo ago
Cool worlds says no? https://youtube.com/shorts/qreL7htXp98?is=G9zJVbpMT0yZsLQ-
32.
by WhitneyLand 3mo ago
The silence is deafening. Google watches over the last few months a flat out assault on the Pareto curve from American and Chinese companies. Release after release pushing the boundaries of frontier intelligence and price/performance.
33.
by WhitneyLand 3mo ago
The world doesn’t need advisors like Moh that are bitter enough to gossip about a family funeral.
34.
by WhitneyLand 3mo ago
Not convinced just using top n sigma is going to beat a SOTA detector. The fingerprints Pangram uses should in principle be able to detect style above the token selection level. Other papers have tried to beat it with temperature and it did
35.
by WhitneyLand 3mo ago
I’m personally interested in it as part of the research to improve LLM writing. Detecting “AI voice” is part of understanding what’s wrong with it in the first place and how to improve it. But yeah, in general I think you’re right, the actu
36.
by WhitneyLand 3mo ago
I’m not sure what you’re saying I’m wrong about. The comment I was responding to, about changing LLM output, referred to prompting, not temp/sampling tricks. I’m not aware of Pangram being beat by clever prompting. There’s some interes
37.
by WhitneyLand 3mo ago
That’s a different point. I’d want detectors to be as accurate as possible, false positives of 1 in 10000 seems like a good starting point. I believe their results have been independently tested. And as a separate matter, any tool for evalu
38.
by WhitneyLand 3mo ago
"It's really easy to have a false positive" Not really. The false positives for the SOTA detector are very very low. "It's also very easy to change the pattern of LLM output." Not in a way that can reliabl
39.
by WhitneyLand 3mo ago
"Text is simply not information dense enough to be able to decode some arbitrary signal of provenance from it...it's a bad fiction to perpetuate that any of this is anything more than tarot card reading." Not true at all. P
40.
by WhitneyLand 3mo ago
I find it hard to see a language as beautiful that’s grown too complex for a single person to hold a complete mental model of. I used to think that was a personal limitation, until I saw an interview with Bjarne explaining that he used to u
41.
by WhitneyLand 3mo ago
Everyone will still need to use Xcode for at least some debugging, no way around that. As for the builds, your agent probably already knows how to do a lot of this from the command line, although explicit suggestions can help it build faste
42.
by WhitneyLand 3mo ago
Your value is intrinsic as a human being. We’re capable of love and shared experiences that a machine will never know.
43.
by WhitneyLand 3mo ago
The prompt does matter. They specifically told it to assume a proof exists so it would not too easily dismiss the possibility.
44.
by WhitneyLand 3mo ago
If all checks out this is a huge milestone. AI has now solved one of the most famous open problems in graph theory, using an off the shelf model, in one hour. It might be a better mathematician than most humans at this point. Kind of like
45.
by WhitneyLand 3mo ago
Codex is supported well on iPhone/iPad, it’s inside the ChatGPT app. It’s amazing how much work you can get done on your phone now, especially if you already have a design mapped out in your head.
46.
by WhitneyLand 3mo ago
That would be cool, but is there a way to differentiate between voice mode and text only output in the custom instructions? Otherwise, it seems like preferences could be competing.
47.
by WhitneyLand 3mo ago
Much better than it was before but it’s still significantly weaker than a direct chat. For example I asked “Why should LLM attention use dot product instead of cosine similarity, being that we often hear vector magnitude does not encode mos
48.
by WhitneyLand 3mo ago
Nice work. It’s advertised 7MB, but also comes with a 5MB mini version. Looks like mini saves space by using 256 element vectors internally instead of 384, but then projects it up to 384 at the end for compatibility. It’s a third smaller, b
49.
by WhitneyLand 3mo ago
“Z.ai provides a replacement MCP for web search, but it's pretty awful and slow” I’ve had good results with Tavily so far, might be worth checking as an alternative for agent search.
50.
by WhitneyLand 3mo ago
This post seems like some decent advice mixed in with a lot of overconfidence and unverifiable claims. “expert developers whose skills have reached the point where they outclass any and all “frontier AI models” in their area of expertise” A
51.
by WhitneyLand 3mo ago
The premise of a secure open codebase is fine. The problem is being more auditable does not automatically make it more audited. There have to be enough people with skill taking enough time to work on it.
52.
by WhitneyLand 3mo ago
What’s with the 3 subscription plans that are suggestive of being mapped to plans from Anthropic and Open AI? Do they really correspond roughly? Seems like they’re trying to suggest a discount while still being worth a significant amount o
53.
by WhitneyLand 3mo ago
This is some high level hard earned wisdom.
54.
by WhitneyLand 3mo ago
How strict is this no AI policy? Say AI is used to identify and rewrite a single function that improves performance or fixes a bug, then the developer carefully reviews and tests it and submits a nice tight PR with all human communication.
55.
by WhitneyLand 3mo ago
Fredrik, while acknowledging everything you said, the fact remains some of my money can end up empowering politics I find repugnant. If it were a small amount of money, it wouldn’t be an issue. If the politics at stake were less important,
56.
by WhitneyLand 3mo ago
I’m not sure how much we’re disagreeing, as I mentioned I don’t claim the gap can’t be closed. I’m saying as of now the gap is real and quantifiable.
57.
by WhitneyLand 3mo ago
That observation doesn’t really prove that the argument fails, right? Just because analogies were used powerfully beyond the experience of a writer doesn’t let us conclude that AI has transcended or will soon transcend the value of all qua
58.
by WhitneyLand 4mo ago
Well image generators work differently… Do you mean that they may get away with less oscillators because of the decoder layer? Well there’s the rub isn’t it, the more work you have done by a software layer the less power you’ve proportiona
59.
by WhitneyLand 4mo ago
Which idea is not outlandish? Physical computing? I agree physical computing is a fascinating topic. But specifically what they’ve simulated here? I don’t see how that would ever work in real life scaled up to any kind of real size. I’m not
60.
by WhitneyLand 4mo ago
It’s not clear to me how this would ever be practical since it seems dependent on n^2 scaling. You’ve got to wonder when you have an image generation demo why would you possibly have 64 x 64 pixel output as your demo? If I’m understanding t

More ›