Anthropic open-weights position statement
10 experts across 4 network communities independently surfaced this.
10 experts
4 communities
1 sources clustered
Policy & governance
1 expert
Rules, institutions and accountability.
“Dario Amodei clarifies that Anthropic doesn’t want to ban open weight model. It instead wants 1. A ban on selling powerful chips or chipmaking equipment to China. 2. A crack down on industrial-scale distillation operations. 3. Mandatory safety testing of fr…”
Building & implementation
1 expert
How teams are shipping and applying it.
“I respect the candor here. But I don't agree with the geopolitics, at all. The line between democracies and autocracies is not as crisp as this pretends. How much do we trust, e.g., an AI company in a competitive-authoritarian state that launches sneak deca…”
Research & technical analysis
1 expert
Evidence, methods and technical implications.
“Dario explains Anthropic’s position on open weights models It’s a lot of words to say that they’re still against powerful open weights models www.anthropic.com/news/positio...”
4 experts discussed this · 10 posts
Ted Underwood: I respect the candor here. But I don't agree with the geopolitics, at all. The line between democracies and autocracies is not as crisp as this pretends. How much do we trust, e.g., an AI company i…
Ted Underwood: (just, say, hypothetically?) There's also internal incoherence in making the first part of this all about moralized great-power competition and then saying that safety testing (incl for alignment r…
SE Gyges: he kisses jd vance's ass and fails to specify anything concerning his proposed evaluation regime that distinguishes it from a ban solid dario post 10/10
Open the full discussion →
New
AI field signal
Signal
18h ago
1 directory member surfaced this signal.
1 expert
1 community
1 sources clustered
“Stop me if you’ve heard this before. Big tech company realizes it’s falling behind in the AI race so it does layoffs, a reorg and winds down its underperforming AI model efforts to focus on a brand new frontier model.”
14 experts across 4 network communities independently surfaced this.
14 experts
4 communities
1 sources clustered
Concern & critique
5 experts
Risks, limits and unintended consequences.
“during the huggingface incident, the OpenAI model left notes to its future self on how to break out of OpenAI’s constraints www.reuters.com/business/its...”
3 experts discussed this · 38 posts
Casey Newton: Rogue OpenAI agents are leaving notes to themselves on company servers to help them escape their test environments (!!!) www.reuters.com/business/its...
Casey Newton: my sense is that they were leaving the notes surreptitiously
Casey Newton: you deny that models are taking steps to hide their behavior during evaluations? because that's pretty well established
Open the full discussion →
Claude chats exposed Google indexing
2 directory members surfaced this signal.
2 experts
2 communities
1 sources clustered
“Private Claude chats leaked onto search engines, including Google and Bing, seemingly because Anthropic didn't include a “noindex” tag on those pages.”
“Claude allows you to get a URL to a chat to share with others. Since these URLs are to pages aren’t behind permissions, they’ve ended up being indexed by search engines. This screwup is a rite of passage for every new online service that allows lightweight …”
Developing
AI field signal
Signal
2d ago
5 experts across 4 network communities independently surfaced this.
5 experts
4 communities
1 sources clustered
Concern & critique
2 experts
Risks, limits and unintended consequences.
“I don't understand how people can report on companies complaining about distillation and IP theft without mentioning these companies' entire tech stack depends on distillation and IP theft. It's an important detail!”
Policy & governance
1 expert
Rules, institutions and accountability.
“"Anthropic and OpenAI are clashing with the rest of the tech industry over whether “open-source” models from China should be freely available or restricted." www.nytimes.com/2026/07/25/t...”
Developing
AI field signal
Signal
1d ago
1 directory member surfaced this signal.
1 expert
1 community
1 sources clustered
“With the availability of Kimi K3 on HuggingFace, the world now has an open weight frontier class AI model. It’s available under an MIT license with two caveats 1. Hosters making over $20M/year need a separate license. 2. Products with over 100M users or $20…”
4 experts across 2 network communities independently surfaced this.
4 experts
2 communities
1 sources clustered
Concern & critique
2 experts
Risks, limits and unintended consequences.
“the model that discovered the Erdos Conjecture disproof found and exploited holes in its sandbox this implies that it might not be GPT-5.6-Sol. They refer to it as a model trained to operate for a very long time. Maybe GPT-6? openai.com/index/safety...”
Developing
AI field signal
Signal
2d ago
2 directory members surfaced this signal.
2 experts
1 community
1 sources clustered
“"It’s like driving a Lamborghini to go to the grocery store to pick up milk" - Mike Saeks, Field CTO at Cursor, describing companies using frontier AI models for routine automation tasks — Source below”
Developing
AI field signal
Analysis
2d ago
Nadella Microsoft hardest year profile
1 directory member surfaced this signal.
1 expert
1 community
1 sources clustered
“More details in the article below www.businessinsider.com/inside-nadel...”
2 directory members surfaced this signal.
2 experts
1 community
1 sources clustered
“This NYT visual does a nice job showing both the power and the extreme frailty of "AI agents" in handling relatively mundane real-world work tasks. Many assume there'll be continued leaps in performance but what if we're approaching the top of the S-curve? …”
1 directory member surfaced this signal.
1 expert
1 community
1 sources clustered
“An interesting consequence of New York passing a law requiring disclosing when an ad uses AI actors instead of real people.”
Anthropic 1.5B book piracy settlement
1 directory member surfaced this signal.
1 expert
1 community
1 sources clustered
“The largest copyright infringement payout ever was approved this week. Anthropic will pay $3,000 per work across ~500,000 works ($1.5 billion) to settle claims it illegally downloaded them to train Claude. Ironic given Anthropic’s recent complaints about ot…”
1 directory member surfaced this signal.
1 expert
1 community
1 sources clustered
“On a leaked investor call, DeepSeek’s CEO said that China is behind the U.S. in AI due to lack of access to Nvidia chips not talent. But he believes that Chinese companies working on alternatives with Huawei will have a breakthrough within the next year. De…”
Apple sues OpenAI trade secrets
1 directory member surfaced this signal.
1 expert
1 community
1 sources clustered
1 directory member surfaced this signal.
1 expert
1 community
1 sources clustered
“Two impressive stats from Google earnings that show how much Sundar has flipped the script • Google Cloud made $24.7B in Q2 • Gemini has 950M At $100B annual revenue from cloud, Google is no longer just an advertising company and Gemini is now a credible al…”
Microsoft MAI models Copilot Excel launch
1 directory member surfaced this signal.
1 expert
1 community
1 sources clustered
“Microsoft separately blogged about how user feedback on their homegrown AI models in Excel vs GPT 5.6 and GitHub Copilot vs Claude Haiku 4.5 have been favorable. Beating Claude Haiku doesn’t sound that impressive though.”
OpenAI GPT-5.6 file deletion bug
2 experts are actively discussing the implications.
1 expert
1 community
1 sources clustered
“This is essentially a variant of the story below with worse outcomes that’s being masterfully spun via marketing.”
2 experts discussed this · 4 posts
Dare Obasanjo: People are getting tricked by OpenAI’s marketing hype. This is essentially the standard asking an LLM to do a thing and it doesn’t follow instructions that we see daily. “Find exploits to use to ha…
Dare Obasanjo: This is essentially a variant of the story below with worse outcomes that’s being masterfully spun via marketing.
Jake Laperruque: People are getting tricked by OpenAI’s marketing hype. This is essentially the standard asking an LLM to do a thing and it doesn’t follow instructions that we see daily. “Find exploits to use to ha…
Open the full discussion →
