U.S. lifts restrictions on Anthropic’s most powerful AI models

How much do we think Anthropic had to pay the Trump family for this? www.nytimes.com/2026/06/30/t...

I love all the endless press coverage of the Anthropic stuff that pretends that the Trump administration has any coherent understanding of anything when it comes to AI policy or guardrails, or that corruption wouldn't immediately disintegrate any useful policy even if it did accidentally materialize

looking at my TL, anthropic finally accomplished what nobody thought any AI lab was capable of: killing a significant chunk of interest in Fable/Mythos

Anthropic is making a strong case against closed source AI models. First they were caught silently degrading Claude’s output if it thinks you’re building a competing product. Now Anthropic admitted having spyware like code that tries to identify Chinese users.

I read an article yesterday saying that Anthropic and OpenAI’s salaries are so high, it’s starting to push “regular”tech workers only making $185k out of San Francisco, which, boohoo, but also seems incredibly unsustainable.

the USG having a stranglehold on AI labs is bad and Anthropic are reaping what they sowed. from here on out no Mythos-class or even Opus-class model will make it out of a US lab without the USG having a piece of the pie. if the plan was to stay ahead of AI risk by leading the pack, well, good luck.

Commerce Secretary Howard Lutnick said the decision was made after Anthropic agreed to “proactively detect and address security risks.”

- De Trump-regering heft de beperkingen op die waren opgelegd aan de nieuwste AI-modellen van Anthropic, Fable 5 en Mythos 5. De modellen waren eerder geblokkeerd vanwege zorgen over nationale veiligheid en mogelijke misbruikrisico’s.

The addition offers new detail to Commerce Secretary Howard Lutnick’s letter announcing the removal of restrictions. “Among other things, Anthropic has agreed to proactively detect and address security risks posed by the models,” wrote Lutnick. Read more: www.wired.com/story/anthro...

Here’s 5: OpenAI, Anthropic, Mistral AI, Cohere, and AI21 Labs. AI’s environmental impact is worth debating. But saying every AI company is “owned and operated by MAGA tech bros” isn’t supported by publicly available evidence.

Anthropic refused to remove safeguards against autonomous weapons and mass surveillance. Trump banned it from federal systems. A federal judge called the ban "Orwellian" and blocked it. The ban was lifted when Chinese AI caught up and America needed Anthropic more than it needed the leverage.

Never a dull day in cybersecurity, so check out today's Metacurity for the most critical developments you should know, including --Anthropic restores Fable 5 and Mythos 5, launches Sonnet 5, 1/5 www.metacurity.com/anthropic-re...

Anthropic being shady af, once again. Combined with their subsidized pricing, it's important to realize - Claude Code isn't the product. You and your data is the product, and Anthropic chooses to upload whatever it wants from your machine. That's why Claude Code isn't open-source.

Fable 5 should be available worldwide again! Anthropic negotiated with the US government and made some changes to the model Will you use it ?! #llm #ai #tech #buildinpublic

U.S. Lifts Restrictions on Anthropic’s Most Powerful A.I. Models www.nytimes.com/2026/06/30/t...