falcor84

30 ms·

541.
by falcor84 9mo ago
You're assuming a world where humans are still needed to read the papers. I'm more worried about a future world where AIs do all of the work of progressing science and humans just become bystanders.
542.
by falcor84 9mo ago
I don't get this argument. Our nervous system is also heterogenous, why wouldn't AGI be based on an "executive functions" AI that manages per-function AIs?
543.
by falcor84 9mo ago
That's one scenario, but I also see a potential scenario where this integration makes it easier to manage the full "chain of evidence" for claimed results, as well as replication studies and discovered issues, in order to the
544.
by falcor84 9mo ago
Reminds me of that dystopian virtual sex scene in Demolition Man (slightly nsfw) - https://youtu.be/E3yARIfDJrY
545.
by falcor84 9mo ago
It seems clear to me that this is about OpenAI getting telemetry and other training data with the intent of having their AI do scientific work independently down the line, and I'm very ambivalent about it.
546.
by falcor84 9mo ago
Sounds like you're in agreement with the parent - outside the US, people see content that reflects poorly on the US, and which is blocked for US citizens
547.
by falcor84 9mo ago
Here's a sad prediction: over the coming few years, AIs will get significantly better at critical evaluation of sources, while humans will get even worse at it.
548.
by falcor84 9mo ago
What does Israel have to do with this article? From what I gather, Iran International is owned by Volant Media, based in the UK, with funding from Saudi Arabia.
549.
by falcor84 9mo ago
Absolutely agreed, we should be more eager to celebrate having tried and lost. One good example that comes to mind is "Such a Loser" by Garfunkel and Oates, with its poignant "It's better to be a loser than a spectator&q
550.
by falcor84 9mo ago
For others as geographically ignorant like me, I just learned that it's in Belgium, a bit east of Brussels.
551.
by falcor84 9mo ago
> AI will not get much better than what we have today Wow, where did this come from? From what just comes to my mind based on recent research, I'd expect at least the following this or next year: * Continuous learning via an archite
552.
by falcor84 9mo ago
Yes. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
553.
by falcor84 9mo ago
It's a question of semantics. What I'm saying is that the way many of us use html in practice in 2026 is less like arbitrary strings and more like db connection strings, where most of our focus is not on whether a bit of text is a
554.
by falcor84 9mo ago
Based on the title, I expected it to say something about this moment in time, using some actual news, but there's nothing there beyond essentially "I don't think this AI think will be useful in the long term".
555.
by falcor84 9mo ago
As Yegge himself would agree, there's likely nothing that is particularly good about this specific architecture, but I think that there's something massive in this as a proof of concept for something bigger beyond the realm of sof
556.
by falcor84 9mo ago
I generally am a fan of polished writing, but I do believe that there's room for quickly fired experimental stuff, and quite enjoyed this piece. With the speed he was going, I wouldn't be surprised if the system architecture actua
557.
by falcor84 9mo ago
I think that it is a debate, and it depends on the role of HTML in your system. If all you're doing is using HTML to "annotate a document with its structure and its formatting", then yes, I'll accept that it's not q
558.
by falcor84 9mo ago
> As intelligence moves into scientific research, drug discovery, energy systems, and financial modeling, new economic models will emerge. Licensing, IP-based agreements, and outcome-based pricing will share in the value created. That is
559.
by falcor84 9mo ago
That's essentially what WASM does, no?
560.
by falcor84 9mo ago
This is not at all clear to me. Reminded of that joke of how "A month in the lab can save you an hour in the library", thinking about some of the best science in history, the researchers often had very strong theory-based belief i
561.
by falcor84 9mo ago
Every single group of people that has been around for long enough accumulates things that they should be ashamed of. But everything is relative, and compared to other countries, Canada and Canadians have always seemed to me to be much bette
562.
by falcor84 9mo ago
As a big nitpick, I found the y axis in that chart ranging between 40%...90% to be horribly misleading.
563.
by falcor84 9mo ago
You can still share via torrents, but Soundcloud seems to be the main place, as it has been for almost 20 years now. On a separate note, I must say that starting to read your comment, I was sure it'll be going towards "Back in 200
564.
by falcor84 9mo ago
Also - Analog Computing [0] seems like it could be particularly well-suited for AI, e.g. by leveraging optics [1]. [0] https://en.wikipedia.org/wiki/Analog_computer [1] https://www.nature.com/articles&#
565.
by falcor84 9mo ago
It's about formalization in Lean, not peer review
566.
by falcor84 9mo ago
I wonder if they could integrate a secondary "world model" trained/fine-tuned on Rollercoaster Tycoon to just do the layout reasoning, and have the main agent offload tasks to it.
567.
by falcor84 9mo ago
It's an interesting slice-of-life piece, and I'll want to watch the film if I have an opportunity, but the site's making this sound unique to Tel Aviv, and the byline's mention of "hypocrisy and cruelty of Israeli s
568.
by falcor84 9mo ago
I agree with the issues, but it definitely doesn't suck if compared to every single other massive endeavor out there. As I see it, it's like that quote about democracy - it's the worst way to attempt to catalogue human knowle
569.
by falcor84 9mo ago
There are agendas there, just like in every human endeavor, but it definitely hasn't been "hijacked", it's still by far the best single repository of human knowledge out there. If I had to choose one website to take with
570.
by falcor84 9mo ago
> to the point of no recovery Interesting, are you saying that the intentional use of USD somehow makes the US more vulnerable? What failure mode are you thinking of?

More ›