Hacker News: adamzenith

New comment by adamzenith in "Neural networks are clean"

adamzenith — Fri, 16 Oct 2026 06:59:16 +0000

One of the most well known interpretability researches released some potentially very useful research as a blog post, and he asked claude to make a cool little thing on top of it?

New comment by adamzenith in "I think I found a planet nobody knew existed. I used Claude Code to find it"

adamzenith — Fri, 09 Oct 2026 10:37:09 +0000

Do you have a source, and also so what? This could have just as easily been codex or any other harness.

New comment by adamzenith in "GPT-6.1 Sol replaces GPT-6 Sol after just 7 days, with near-Astra intelligence"

adamzenith — Wed, 30 Sep 2026 11:14:51 +0000

You don't think having a more intelligent model they can use internally that others can't is an advantage?

New comment by adamzenith in "Why do we need human mathematicians anymore?"

adamzenith — Mon, 21 Sep 2026 11:12:24 +0000

> There is nothing to suggest that an AGI will care about its future or who controls it.

Isn't this exactly the scary thing? A sufficiently advanced agent unleashed with an imperfectly defined goal will do whatever it determines it needs to do to accomplish it, careless of the consequences.

New comment by adamzenith in "DeepSeek v4.1 Flash"

adamzenith — Thu, 10 Sep 2026 09:12:26 +0000

The field is much older, MIRI is ~20 years old. Look up Eliezer Yudkowsky.

New comment by adamzenith in "Note-Taking and Personal Knowledge Management"

adamzenith — Mon, 03 Aug 2026 10:25:42 +0000

I think more the idea that wrestling with new ideas and distilling them down can you help you to understand something, even if that note is then immediately thrown out. The purpose of the notetaking in this case is not for future reference.

New comment by adamzenith in "Some thoughts about Anthropic's new cryptanalysis results"

adamzenith — Wed, 29 Jul 2026 18:42:07 +0000

In response to your edit, you should check out Terry tao's chat gpt logs about the recent Jacobian result. The models are smart enough to brute force some things, but can cut to the meat much faster with good prompting

New comment by adamzenith in "Pacing the frontier"

adamzenith — Wed, 29 Jul 2026 17:56:49 +0000

Not under all the competitive pressure they are facing, but if an industry wide slowdown could be achieved they might spend more time on model alignment and less on pure capabilities.

New comment by adamzenith in "The CEO of Mullvad is the main financer of the Swedish Örebro party"

adamzenith — Tue, 30 Jun 2026 08:48:00 +0000

What is the issue with Proton?