Hacker News: bavell

New comment by bavell in "Cloud Agents Are Inevitable AI Prisons"

bavell — Wed, 23 Sep 2026 20:45:12 +0000

How long did it take these companies to even notice? Why wasn't exploiting bugs in the agent sandboxes anticipated?

Human failures all around, though it's easier to just blame the models.

New comment by bavell in "Government Rails Site Hit Hours After CVE Patch"

bavell — Sat, 05 Sep 2026 02:52:11 +0000

Android Firefox here, top nav takes up 1/3rd of the viewport

New comment by bavell in "Ask HN: Who is using MCP in production?"

bavell — Fri, 04 Sep 2026 19:54:07 +0000

It's a bit of a nothingburger IMO, but charitably you could think of it as a rest API + bundled agent instructions.

New comment by bavell in "Linux 7.3 improves performance when running out of vRAM"

bavell — Tue, 18 Aug 2026 13:13:04 +0000

I use Arch exclusively and haven't had any of the issues you describe.

That said, I do occasionally deal with quirks due to the rolling distro nature of Arch. The benefits far outweigh the occasional upgrade pain IMO though.

New comment by bavell in "Ask HN: What are you working on? (August 2026)"

bavell — Mon, 10 Aug 2026 00:09:02 +0000

What was/is it like working with Hooded Horse? How did you get that going? I'm a big fan of the games they publish :)

Good luck on your game & launch!

New comment by bavell in "Kimi-K3 on HuggingFace"

bavell — Mon, 27 Jul 2026 12:03:57 +0000

I find it difficult to always have one or more long-horizon tasks 'queued up' and ready to run... I find myself usually bottlenecked on design, review, or something similar that requires me being in the driver's seat. It's possible I could queue up a bunch of tasks, letting the LLM run off in multiple directions, but then I'd be less able to steer and course correct.

Just my experience though, I'm still figuring things out. Perhaps some subsets of tasks would be more ideal for these long-horizon workloads - exploration, multiple competing implementations, etc...

New comment by bavell in "A taxonomy of omnicidal futures involving artificial intelligence (2025)"

bavell — Fri, 24 Jul 2026 01:19:37 +0000

I find no comfort in it. Merely pointing out that 'some' is hand-wavy and perhaps technically true but meaningless to the discussion.

New comment by bavell in "A Taxonomy of Omnicidal Futures Involving Artificial Intelligence"

bavell — Fri, 24 Jul 2026 01:18:04 +0000

Precisely

New comment by bavell in "Why Software Factories Fail (or: harness engineering is not enough)"

bavell — Fri, 24 Jul 2026 00:58:23 +0000

Good read and matches my experience pretty well, but I'd humbly suggest the author look up the definition of 'vertical' and 'horizontal' and swap their terminology + fix their videos :)

New comment by bavell in "A taxonomy of omnicidal futures involving artificial intelligence (2025)"

bavell — Fri, 24 Jul 2026 00:09:39 +0000

'Some' is doing a lot of heavy lifting there...

New comment by bavell in "Ask HN: Did Fable disappear from your Claude usage and requires credits now?"

bavell — Fri, 17 Jul 2026 18:44:54 +0000

Wouldn't be surprised if they issued credits or extended the deadline again as an apology.

  > Update - We are aware of an issue preventing users from selecting Claude Fable 5 within Claude.ai, Claude Code, and other surfaces, and are working to resolve this issue.

https://status.claude.com/

New comment by bavell in "Ask HN: Did Fable disappear from your Claude usage and requires credits now?"

bavell — Fri, 17 Jul 2026 18:27:47 +0000

Same here. Weirdly,

> The promotion starts July 1, 2026 and ends July 19, 2026 at 11:59:59 PM PT.

from: https://support.claude.com/en/articles/15424964-claude-fable...

New comment by bavell in "AI can't be listed as inventor on patent applications, Japan's top court rules"

bavell — Thu, 02 Jul 2026 15:08:44 +0000

> "... the plaintiff submitted an application in 2020 for food containers and other items invented by DABUS, an artificial intelligence the plaintiff had created."

The plantiff is Stephen Thaler: https://imagination-engines.com/founder.html

New comment by bavell in "Claude Fable 5 Promotional Access"

bavell — Wed, 01 Jul 2026 21:27:05 +0000

Fable was only available for a few days, not seven.

New comment by bavell in "Extreme Heat conference cancelled due to extreme heat warning"

bavell — Wed, 24 Jun 2026 02:17:16 +0000

Yep, 9:30p here and it's 82F/80% humidity. Still pretty mild compared to the deep summer months (Jul/Aug)!

New comment by bavell in "Beyond All Reason (Free Total Annihilation Inspired RTS)"

bavell — Sun, 21 Jun 2026 15:14:46 +0000

Not my experience at all when I played around a year ago. There were plenty of noob-specific / noob-friendly lobbies with a mix of helpful players and fellow casuals.

New comment by bavell in "U.S. science is in chaos"

bavell — Thu, 18 Jun 2026 00:59:04 +0000

My rock collection causes concepts to enter my brain, but I don't think I'd say they're communicating with me, nor I with them.

New comment by bavell in "Ask HN: What was your "oh shit" moment with GenAI?"

bavell — Sat, 06 Jun 2026 13:14:54 +0000

Another option - we keep the wild west unverified web, dotted with islands of verified/vetted spaces.

New comment by bavell in "Entanglement Builds Space-Time. Now "Magic" Gives It Gravity"

bavell — Fri, 05 Jun 2026 14:52:44 +0000

This was a helpful visualization: https://youtu.be/U_sI9agWmEw?si=MItDfnTx1-oT_qX_

New comment by bavell in "I'm Tired of Talking to AI"

bavell — Wed, 27 May 2026 13:24:32 +0000

> after I realize what's happening, I tend to take a little longer for each reply so they figure out it's faster to just do the research on their own most times.

Agreed, and I do the same. They still get a courteous reply, but they also feel a little "pain" when they don't get a timely answer - an effective teacher.