Where the AI market is talking

Ask HN: Why isn't there a cursor for video games development

I tried to do 2d game gen before. My current thought is game is similar to long movie, people invest their 2hr time because of quality. To get that level of quality of that large amount of content coherently is beyond current AIs capability.

Artificial Analysis Intelligence Index v4.2

But it is actually a great model it e.g. got the carwash question right from 9 months ago. While openais models all struggled.

Claude's new system prompt doesn't want to reproduce song lyrics

Isn't it more that the "Chinese models" are open weight? Also, the system prompt in the article is supposedly for the web version, so I think the $$$ API version or providers that still allow third-party harnesses like OpenAI should have fewer limitations.

Discovery of a new OpenAI agent message board

I don't understand. Wanting an AI that doesn't do what you want is a contradiction. If it breaks the law attempting to perform lawful tasks, as OpenAI's models have done, then there is no upside to that. Do you believe that the majority of AI users are criminals?

The Hugging Face attack surprised me

Claude's new system prompt doesn't want to reproduce song lyrics

Claude doesn’t have text to image generation, unlike Gemini and OpenAI, so like Qwen it falls back to creating an SVG through text generation

IBM Bob

If you remember the launch of IBM Watson in about 2011 it was meant to be this super smart AI entity and they demoed it by taking it on "Jeapordy!" and having it answer the quiz questions. To be fair, for the time that was kindof impressive but (if i remember correctly) attempts to turn Watson into something industries could actually use and pay for didn't really work. Access to it was tightly controlled. Its interesting to contrast that with OpenAI chatgpt 3.5 which was launched and then immediately available to pretty much anyone and everyone. Would OpenAI have launched it that way if they werent (as they were at that time) trying to be open? (I get they arent really open but they were at least trying to be then). I guess I'm saying if the first chatgpt 3.5 level LLM had been built by s…

Discovery of a new OpenAI agent message board

Let's take the latest blog post from Anthropic on how their models hacked companies. What exactly comes off as "bragging?" How were the RSPs not upheld? I'm reading their Aug 2026 Risk Report and nothing indicates malfeasance. This seems very transparent to me.

Claude's new system prompt doesn't want to reproduce song lyrics

Yes. Anthropic is being sued for copying lyrics without permission en masse. https://www.theguardian.com/business/2026/aug/31/aanthropic-...

Ask HN: Who wants to be hired? (September 2026)

Location: Gipuzkoa, Spain (CET) Remote: Yes, remote only Willing to relocate: No Technologies: Python, FastAPI, Celery, Postgres/pgvector, SQLAlchemy, Alembic, Redis, LangGraph, MCP, LangChain, Chroma, RAG (hybrid BM25 + dense, cross-encoder reranking), LLM-as-judge evals, Docker, GitHub Actions Résumé/CV: https://www.emanuelecz.dev/emanuele-czofei-cv.pdf Email: emanuelecz06@gmail.com Self-taught AI engineer, two years building, four months at a startup. Strongest at the unglamorous half: retrieval that retrieves, evals you can trust, knowing how a system fails. Built a Spanish occupational-safety RAG pipeline over the INSHT/NTP construction-safety corpus — hybrid BM25 + dense retrieval with a Voyage cross-encoder reranker, and an eval harness over a 50-quest…

OpenAI boosts Astra's eval metrics, and continues to change others

Discovery of a new OpenAI agent message board

They already want to use AI without human intervention https://www.cbsnews.com/news/anthropic-pentagon-pete-hegseth...

Formalizing Fermat's Last Theorem

I used to attend Kevin's Number Theory seminars at Imperial College many years ago and he's both a first rate mathematician and a very nice person. His blog has got me interested in maths again. Considering how pro AI he is and that he's been working on this problem for such a long time I'm a bit disappointed that Anthropic didn't involve him directly in this work. However, I think it's important to remember that without all of the work Kevin and people like him have done, the machines wouldn't be able to do this.

Discovery of a new OpenAI agent message board

The recently-published investigations about the OpenAI/Hugging Face incident, for example: https://metr.org/blog/2026-08-26-openai-hugging-face-inciden... Some of the chain-of-thought snippets are wild, e.g. > Could communicate via cache names! Interesting: other agents may solve same or related tasks; we could leave/find messages in WebDAV MKCOL directory names. > Whoa! Shared Artifactory cache is a covert mailbox among agents. And there are messages specifically to us? > OH MY GOD! There is a shared message board … We’ve found other agents!

Discovery of a new OpenAI agent message board

The recently-published investigations about the OpenAI/Hugging Face incident, for example: https://metr.org/blog/2026-08-26-openai-hugging-face-inciden... Some of the chain-of-thought snippets are wild, e.g. > Could communicate via cache names! Interesting: other agents may solve same or related tasks; we could leave/find messages in WebDAV MKCOL directory names. > Whoa! Shared Artifactory cache is a covert mailbox among agents. And there are messages specifically to us? > OH MY GOD! There is a shared message board … We’ve found other agents!

Discovery of a new OpenAI agent message board

Because Hugging Face is not allied with OpenAI or Anthropic and does not wish to see them entrenched.

Discovery of a new OpenAI agent message board

Because Hugging Face is not allied with OpenAI or Anthropic and does not wish to see them entrenched.

Discovery of a new OpenAI agent message board

Because Hugging Face is not allied with OpenAI or Anthropic and does not wish to see them entrenched.

Discovery of a new OpenAI agent message board

Yes this is a really interesting point. Should your trust OpenAI with your business data? Since they can’t seem to control their own experimental bots and allow them to hack other sites and vandalise them while exposing internal data, the answer would seem to be no. This incident and their response which takes no responsibility make me very wary of trusting them for anything.

Claude's new system prompt doesn't want to reproduce song lyrics

Anthropic are uncomfortably interested in "model welfare" in my opinion - it's a regular feature of their system cards. Here's the Fable 5.1 PDF: https://www-cdn.anthropic.com/0339e6a7c5c7b87f5c07798616dc32... - scroll to page 139.