avaer

23 ms·

451.
by avaer 9mo ago
I guess the problem is one of legal attribution. If a human takes responsibility for the AI's actions you can blame the human. If the AI is a legal person you could punish the AI (perhaps by turning it off). That's the mode of res
452.
by avaer 9mo ago
I 100% agree. I'm just trying to point out the problem isn't Microsoft AI slopping their software. Even if you slopped it, the software could turn out better than what they're putting out. There must be something much worse t
453.
by avaer 9mo ago
You can literally one-shot Opus 4.6 to make a better, faster, safer, more secure notepad.exe than the one that comes with Windows. This isn't an AI slop problem.
454.
by avaer 9mo ago
I would expect a 1000+ karma account to know better than to shitpost AI slop on hacker news. (if this is not slop, I don't know what to say... the only way your comment relates to the article is the words "Blade Runner")
455.
by avaer 9mo ago
If anyone is curious I implemented a prototype of this at https://github.com/avaer/markdownlang I used Opus 4.6. In my defense, in addition to wanting to try a live version of this because I think it's super cool,
456.
by avaer 9mo ago
Requiring id to get access to a model is one issue. Pulling a switcheroo on the user behind the scenes, whatever the justification, is another issue, and I think the more interesting one. It's a stepping stone to "we will reconfig
457.
by avaer 9mo ago
When the GPT-5 router architecture was introduced I worried that OpenAI would use the technology as a pretext to mislead or defraud users by substituting in worse quality when they could get away with it and then "blame it on the AI&qu
458.
by avaer 9mo ago
There's a famous XKCD about this: https://xkcd.com/2347/ In this case the hero's name is apparently Simon Josefsson (maintainer).
459.
by avaer 9mo ago
Since I didn't see it in the Readme, how does this compare to something like Google's A2UI? Seems like it's doing more, but could e.g. Tambo work on top of A2UI protocol or is it a different beast? My agents need a UI and I&#
460.
by avaer 9mo ago
The hidden fallacy in your comment is that there is such a thing as "news that is really relevant for you". This isn't all that different than saying that it would be nice if someone else did your thinking for you -- which is
461.
by avaer 9mo ago
The repo does make a case for this, namely speed, which does make sense.
462.
by avaer 9mo ago
This feels like the time I was a Mercurial user before I moved to Git. Everyone was using git for reasons to me that seemed bandwagon-y, when Mercurial just had such a better UX and mental model to me. Now, everyone is writing agent `exec`s
463.
by avaer 9mo ago
You were downvoted but I don't understand why. This is the purpose/spirit of antitrust law [1] [1] https://en.wikipedia.org/wiki/United_States_antitrust_law
464.
by avaer 9mo ago
I wish I could agree with you, but as a game dev, shader author, and occasional asm hacker, I still think AIs have demonstrated being perfectly capable of copying "those effects". It's been trained on them, of course. You&#x2
465.
by avaer 9mo ago
Rest assured that when/if this becomes possible, the model will not be available to you. Why would big AI leave that kind of money on the table?
466.
by avaer 9mo ago
It's already in Cursor. I see it and I didn't even restart.
467.
by avaer 9mo ago
You're right, but the people that say they're pilled/sticking it to the man/breaking out of the matrix -- they're just parroting things they saw on X or tiktok. And then shuffling their money based on that. They hav
468.
by avaer 9mo ago
To a first approximation ~nobody who "invests" in crypto has read a whitepaper or considered the ethos. It's one of the bugs/features of crypto.
469.
by avaer 9mo ago
I wish more people were as astute. The front matter of big politics and big tech is not too different from professional wrestling.
470.
by avaer 9mo ago
> An AI agent that clicks buttons. Are people's agents actually clicking buttons (visual computer use) or is this just a metaphor? I'm not asking if CU exists, but rather is this literally the driver of people's workflows?
471.
by avaer 9mo ago
I was really not expecting Apple to jump on this bandwagon, but I guess this was inevitable.
472.
by avaer 9mo ago
> IMHO AI generated content should be treated the same way with how human generated content and I don't see the problem. The problem is that AIs don't have rights. So you literally can't treat it the same. If I "make&
473.
by avaer 9mo ago
I don't think that was the "tech circles" being pro-copyright though. I've never met someone who wasn't on Aaron's side on that one.
474.
by avaer 9mo ago
Don't worry, they will flip and become staunch supporters of AI too once it starts to benefit them. Then they will immediately forget the copyright issue.
475.
by avaer 9mo ago
You might be looking at a small time horizon. What about https://en.wikipedia.org/wiki/An_Open_Letter_to_Hobbyists ?
476.
by avaer 9mo ago
It's not about instructions, it's about discoverability and data. Yeah, WWW is really just text but that doesn't mean you don't need HTTP + HTML and a browser/search engine. Skills is just that, but for agent capabi
477.
by avaer 9mo ago
I agree; this project looks impressive, but I'm guessing there are some rough edges in the transpilation "magic" that should be called out. That's the crux of how usable this is going to be for people's use cases, a
478.
by avaer 9mo ago
I increasingly mentally translate to "authored by Claude but this person claims the credit". Which I hope is not future facing, but maybe that is the future we are facing.
479.
by avaer 9mo ago
Quick Start git clone https://github.com/anthropics/nanoclaw.git Is this an official Anthropic project? Because that repo doesn't exist. Or is this just so hastily thrown together that the Quick Start is a
480.
by avaer 9mo ago
The links have been posted to HN if you search. https://moltroad.com/ comes to mind. The "top rated" on there describes itself as "trading in neural contraband". That's in addition to all of the act

More ›