good morning. In case you missed last night’s breaking news, a federal judge blocked the Trump administration’s ban on Anthropic, ruling that the Department of Defense’s supply chain risk designation for Anthropic’s AI technology was “Orwellian” and punitive. Although this is just a temporary pause while lawyers on each side prepare for the main battle, the judge’s choice of words clearly does not bode well for the government’s case.
luck
Jeff John Roberts’ articles on prediction markets, “Calci” and “Polymarket,” won an award in the Investment/Markets category from Medium Publications. The judges called the book an “entertaining read” with “impressive sources” and “lucid thinking” about the battles on the front lines of an emerging industry.
Additionally, Jessica Matthews and Leo Schwartz’s investigation into Elon Musk’s Boring Company received an Honorable Mention in the transportation category for its “powerful reporting on the risks faced by those building and maintaining tunnels beneath Las Vegas, the $400,000 extinguished fine, and the cozy relationship between the company and state authorities.”
Today’s news is below.
Alexei Oreskovitch
@lexnfx
Want to send thoughts or suggestions to Fortune Tech? Type a line here.
Was this newsletter forwarded to you?Subscribe here!
Developed by AI company Anthropic Following the data breach that revealed the model’s existence, the company said it has begun testing with early access customers a new AI model that is more powerful than any previously released AI model.
An Anthropic spokesperson said the new model represents a “step change” in AI performance and is “the most capable we’ve ever built.” The company said the model is currently being trialled by “early access customers.”
A model description was accidentally stored in a publicly accessible data cache and reviewed by. luck.
A draft blog post published on an insecure, publicly searchable data store by Thursday evening calls the new model “Claude Mythos” and says the company believes it poses an unprecedented cybersecurity risk.
The same cache of insecure, publicly discoverable documents revealed details of a planned invitation-only CEO summit in Europe as part of the company’s efforts to sell AI models to large enterprise customers.
In response to questions about a draft blog post, the company confirmed that it had trained and tested the new model. “We are developing a general-purpose model that has made meaningful advances in inference, coding, and cybersecurity,” an Anthropic spokesperson said. “Given the strength of its capabilities, we are thinking carefully about how we release it. As is standard practice across the industry, we are working with a small group of early access customers to test the model. We believe this model is a step change and the most high-performing we have ever built.” –
Apple is making major changes to Siri that will allow users to connect its aging virtual assistant (first released in 2011!) to other companies’ AI tools. bloomberg.
Changes to Siri that could be rolled out in June as part of Apple’s new iOS 27 software will allow users to tap into third-party AI apps like Anthropic’s Claude and Google’s Gemini to perform certain tasks or answer queries. Apple currently routes many such Siri requests through OpenAI’s ChatGPT as a result of a partnership that Apple and OpenAI announced in 2024.
The iPhone maker is famous for its tight control over the core functionality of its platform, but Apple has struggled to develop its own in-house AI tools, forcing the company to swallow its pride and make concessions. Apple also announced earlier this year that it was partnering with Google to provide some of the iPhone’s core AI capabilities, known as Apple Intelligence. That partnership is separate from the planned Siri changes reported by bloomberg
blue
financial times. The decision reportedly follows concerns among employees and investors about the potential social implications of sexualized AI, including the risk of users forming unhealthy emotional bonds and the potential for minors to access explicit content. The company confirmed that it plans to pause its chatbot deployment, saying it will prioritize long-term research into how people interact with sexually explicit AI before deciding whether to proceed.
The move reflects a broader shift in the company’s strategy, with OpenAI pulling back from projects like video model Sora and focusing on core products like ChatGPT and developer tools.
The proposed “adult mode” is reportedly already facing various technical hurdles, particularly when it comes to safely training models to handle explicit content while filtering out illegal or harmful content. There were also questions over age verification systems, which are not fully reliable.—beatrice nolan
Brandon Tseng, co-founder and president of AI-powered drone maker ShieldAI, said global conflicts are awakening governments and investors to the importance of modernizing military forces. The company announced Thursday that it has raised $1.5 billion in Series G funding at a valuation of $12.7 billion.
The new funding more than doubles the valuation of the San Diego, Calif., defense startup founded in 2015 and previously valued by investors at $5.6 billion. Shield AI is projecting more than 80% revenue growth by the end of 2026, Shield AI cofounder Brandon Tseng and CFO Kingsley Afemikhe told luck. That would equate to at least $540 million in revenue this year, based on Shield AI’s 2025 revenue figures.
“We don’t expect growth to slow down,” Tseng said in an interview. —jessica matthews
More technology
— The IPO is scheduled for June.
—Netflix raises prices. Also.
—
—New tricks for WhatsApp.
—. Basic copy editing only.
—. See you in October.
