xrd

40 ms·

421.
by xrd 1y ago
A friend asked me whether I was concerned about energy usage of AI. I didn't have a good answer. It feels inevitable. But, I love the write-up here on why the sailboat, and why UXN, because those two things are complementary when you a
422.
by xrd 1y ago
I've been worried about the US abandoning Japan and this radically shifting the power centers in the world. But, the way you describe it sounds like I should not worry. At least, as long as I trust the weapons manufacturers to keep thi
423.
by xrd 1y ago
Surrender by whom? Isn't his argument that leaders in the AI/ML space have consistently dismissed the need for that the entire point of the article? And that seems like a valid question to be after reading it. And huge financial i
424.
by xrd 1y ago
This is terrific. I'm a huge fan of Ev and hope medium succeeds. I'm surprised there aren't more investors clamboring to get in now that substack has proven there is a model for good writing and good compensation.
425.
by xrd 1y ago
I have been reading HN for years. The exploits used to be so clever and incredible feats of engineering. LLM exploits are the equivalent of "write a prompt that can trick a toddler."
426.
by xrd 1y ago
Damn, this is brilliant. Thank you.
427.
by xrd 1y ago
I was hoping I would goad someone into doing it. But, do it for emacs, ok? </joke> Actually, I *do* prefer emacs.
428.
by xrd 1y ago
Isn't it more appropriate to compare this to aider? I prefer the command line tools to IDE integration, even though I don't feel like the contextual options are great. In other words, I don't always feel that I can see the ch
429.
by xrd 1y ago
And, who also happened to invent the comic art form called manga. https://en.wikipedia.org/wiki/Hokusai_Manga
430.
by xrd 1y ago
You put it so well! I agree wholeheartedly. Llms are language toys we get to play and it's so much fun. But I'm bothered in the same way you are and that's fine.
431.
by xrd 1y ago
Is no one else bothered by that way of using tools? Tools feel like a way to get deterministic behavior from a very hallucinatory process. But unless you put a very lengthy and comprehensive non-deterministic English statement, you can&#x27
432.
by xrd 1y ago
Arrested in a men's bathroom, no less!
433.
by xrd 1y ago
How do you do this in a remote environment?
434.
by xrd 1y ago
I worry the next step will be to create a social media site for parrots and then they will be just as lonely and angry as we humans.
435.
by xrd 1y ago
If you change the target branch it will only show commits that are against that. I wish GitHub did that automatically because it has the parent commit tree but when do you that you'll only see commits that diverge from the target branc
436.
by xrd 1y ago
I really want this, but BHVS, where react is replaced with svelte.
437.
by xrd 1y ago
I'm scratching my head a bit at this one. I already assume that the models are shifting underneath me. It's very frustrating that most non-developers just think you can ask an LLM a question and it will respond accurately each tim
438.
by xrd 1y ago
I'm really fascinated by this. This looks to me like a great way to log claude code interactions. Is it just me that cannot figure out a way to log the prompts and responses when using Claude Code? If I look here I see nothing about a
439.
Ambsheets(inkandswitch.com)
6 points by xrd 1y ago|2 comments
440.
by xrd 2y ago
It's fantastic. I just watched Kitsune, thanks for sharing. It reminds me why Flow was so good. Flow was great because I could see the shader artifacts. It was the opposite of a Disney model, it was not polished and perfect. That'
441.
MCP Run Python(github.com)
173 points by xrd 2y ago|66 comments
442.
by xrd 2y ago
Wrong code on the Internet does not steal your credit card information. Wrong code on localhost does. But, I think we agree, anyway.
443.
by xrd 2y ago
Good point. And, surely a naughty LLM will get hacked and say "Hey, you've heard about this great thing called linting. Let me configure the newest and bestest one for you, it's called Rm-RF-Star..."
444.
by xrd 2y ago
I'm saying no hallucination will be happening. The LLM will happily write code that permits network access, because it read online an example that did that. And, unless you know better, you won't know to manually turn that off. Sa
445.
by xrd 2y ago
The only real solution I see is lint and ci tooling that prevents non approved packages from getting into your repo. Even with this there is potential for theft on localhost. There are a dozen new YC startups visible in just those two sente
446.
by xrd 2y ago
It doesn't seem like a stretch that an LLM will very soon be able to configure your dependent web assembly components to permit the dangerous access. It feels like this model of security, while definitely a step in the right direction,
447.
by xrd 2y ago
I did not know about Kalshi until today. A blog I really like pointed me to this discussion: https://moontower.substack.com/p/dire-wolf Also, I really like Michael Lewis's podcast about the rise of sports gambling
448.
by xrd 2y ago
I learned it first on HN a few weeks ago, but someone else said it was 4 casinos.
449.
by xrd 2y ago
[flagged]
450.
by xrd 2y ago
Does it have parental controls, or a way of at least seeing what the kids are watching? I can't tell from looking at that page.

More ›