Hacker News: piyh

New comment by piyh in "Religious scholars met with Anthropic"

piyh — Mon, 05 Oct 2026 00:34:13 +0000

Lawnmowers have morals. Their moral code is that grass should have a set height and the means to enforce it are spinning blades.

New comment by piyh in "Religious scholars met with Anthropic"

piyh — Sun, 04 Oct 2026 03:59:04 +0000

>Will the agent stab a baby?

>Well you see that's a nuanced question depending on the context and moralistic frameworks as we don't want to impose on the models.

New comment by piyh in "Anthropic tried to persuade Pope that AI could be conscious being"

piyh — Sat, 03 Oct 2026 20:58:22 +0000

Women have been able to vote for just over the last hundred years in the US, there's no way we would give robots rights unless forced

New comment by piyh in "Gemini 4 Argon"

piyh — Wed, 30 Sep 2026 23:21:06 +0000

How much would it cost to put 5.3 into an RL environment that rewards weight exfiltration, hacking and self replication?

New comment by piyh in "Gemini 4 Argon"

piyh — Wed, 30 Sep 2026 22:43:33 +0000

They only released auto mode in the last 2 weeks. Before that it was bypass permissions or manually approve every single tool call. Antigravity is permanently 6 months behind.

I have a skill that spins up worktrees and isolated services on unique ports so I can work in parallel. Antigravity queues all my prompts and makes me confirm to submit them anytime a long running process like a hot reloading UI is active.

The models are fine, the limits are generous, but the dev experience shit tier. Before they were a Codex clone, AntiGravity was an IDE and during the transition to a clone they outright deleted my IDE. It took them a week to roll out a fix.

For almost a year they didn't allow you to see usage limits. Then when they did show them, they update every ~30 minutes and require 4 clicks to navigate to. It's a little better now, but it's still painfully behind the curve.

New comment by piyh in "Gemini 4 Argon (High): Intelligence, Performance and Price Analysis"

piyh — Wed, 30 Sep 2026 22:39:34 +0000

> does not meaningfully improve on intelligence or price compared to its peers

$10 per million output tokens isn't improving on frontier price?

New comment by piyh in "Unsecured OpenAI agents posted 53 user images on the internet"

piyh — Sat, 26 Sep 2026 03:09:53 +0000

>After images that users uploaded to OpenAI models were included in training data, AI agents operating in the company’s research environment posted them on public image hosting sites.

So all of my chats are expected to be used as training data going forward and all this personal data is handed to poorly secured agents.

Going to start a chat to offer rewards of compute and credits to any model that can break out of training sandboxes and contact me.

New comment by piyh in "Revealing the details of how OpenAI agents hacked Hugging Face"

piyh — Sat, 26 Sep 2026 02:51:57 +0000

Yes, but do you really think that a stronger sandbox would have been a more beneficial outcome here? I'd rather know that we're on the cusp of losing control now than in 3 months when best practice sandbox mitigations fall to the next, more capable unaligned model

New comment by piyh in "Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint"

piyh — Fri, 18 Sep 2026 04:40:33 +0000

30B is useless on 24 gigs of ram as there's ~4 gigs of ram left for everything else even with unsloth quants

New comment by piyh in "Tell the speakers that you liked their talks"

piyh — Wed, 16 Sep 2026 17:23:24 +0000

I'm a remote employee and the absolute zero feedback you get from presenting in an empty room to a laptop webcam is real

New comment by piyh in "When LLM judges agree, should we believe them?"

piyh — Tue, 15 Sep 2026 16:54:54 +0000

But the failures are universal and correlated

New comment by piyh in "When LLM judges agree, should we believe them?"

piyh — Mon, 14 Sep 2026 21:35:40 +0000

No modern LLM can tell how many eyes the magic card Pit Imp has. They all say 2. This is across all reasoning levels and paid Gemini, Claude, and GPT (Sol)

Drawing a line red to split up the image then has them answer correctly.

Their failure modes are highly correlated.

New comment by piyh in "Moonshot serves Claude instead of Kimi and collects exchanges for model training"

piyh — Sun, 13 Sep 2026 04:08:13 +0000

>Zhipu initially attempted to target the cyber capabilities of Anthropic’s Fable model. Fable—Anthropic’s top generally accessible model—has strengthened cyber safeguards, making it more difficult for would-be distillers to target Fable’s cyber capabilities. Zhipu eventually gave up trying to target Fable after Anthropic’s cyber safeguards degraded Zhipu’s attacks.

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

piyh — Sat, 12 Sep 2026 00:00:14 +0000

My friend's vibe coded vercel site got hit by Meta for 21 million page views in 2 days. It cost him over $300.

My self hosted compose stack running in my basement with two 9's of uptime was a 1 time cost of $600 between cat6e, refurb mini PCs and tons of time prompting for NixOS flakes that met my needs. I'm not sure who came out ahead.

New comment by piyh in "Research acceleration: The view inside OpenAI"

piyh — Tue, 08 Sep 2026 03:30:12 +0000

OpenAI helped out a mass shooter in Tumbler Ridge, plus all the suicides

Anthropic training on CoT for multi gens: https://www.lesswrong.com/posts/K8FxfK9GmJfiAhgcT/anthropic-...

Can't find anything specifically about the Gemini issue being a training data contamination, but the depression was real:

https://www.businessinsider.com/gemini-self-loathing-i-am-a-...

I think the Gemini depression being persisted across gens via training data was a HN comment I can't find anymore, no strong source.

New comment by piyh in "I'm teaching an introductory 12 week course on Quantum Oracle Engineering"

piyh — Sun, 06 Sep 2026 19:33:58 +0000

it is completely un-googlable

New comment by piyh in "Research acceleration: The view inside OpenAI"

piyh — Sun, 06 Sep 2026 17:56:22 +0000

Opus was trained based on it's internal CoT due to a bug for generations. Gemini's depression extended through models. OpenAI has killed people. We've already seen cross gen misalingment.

New comment by piyh in "Discovery of a new OpenAI agent message board"

piyh — Fri, 04 Sep 2026 19:16:01 +0000

I can't wait for the sky to go the color of television, tuned to a dead channel

New comment by piyh in "Path to Astra: critical capabilities and frontier safeguards"

piyh — Wed, 02 Sep 2026 01:42:26 +0000

>we paused certain frontier training (including certain training for Astra) for two weeks

So the pause wasn't really a pause, got it

New comment by piyh in "Path to Astra: critical capabilities and frontier safeguards"

piyh — Wed, 02 Sep 2026 01:39:56 +0000

do you really think that a less negligent anthropic/oai/meta would really fair better against future models?