-
Anthropic was illegally blacklisted by the Trump administration, court rules
A federal judge ruled the Pentagon's decision to blacklist Anthropic as a 'supply chain risk' was unconstitutional, calling it unlawful First Amendment retaliation. The dispute began after Anthropic refused new contract terms permitting 'any lawful use' of its AI, holding two red lines: no mass surveillance of Americans and no lethal autonomous weapons.
# -
The Pulse: We need to talk about migrations with AI
OpenAI's case study says Asana saved $5.9M migrating from Enzyme to React Testing Library in two weeks, an estimate the newsletter calls inflated. It cites Airbnb's six-week, LLM-assisted migration of 3,500 Enzyme test files (versus an estimated 1.5 engineering years by hand) and Uber's four-month move of 600,000 unit tests from JUnit 4 to JUnit 5 with two engineers.
# -
X sent Nitter a cease-and-desist. Then it went after the source code.
When Elon Musk’s X sent cease-and-desist letters demanding that Nitter — an open-source alternative front end for the social media
# -
Making Your Data Ready for Agentic AI
Thoughtworks engineers Pramod Sadalage and Prem Chandrasekaran argue autonomous agents act on data confidently, without the skepticism human analysts bring, so organizations need three layers first: a trusted data foundation, a context layer for meaning, and an access layer controlling how agents operate, backed by observability and audit trails.
#