Andy Jassy told Treasury Sec. Scott Bessent that Amazon researchers used Anthropic’s Claude Fable 5 to obtain info usable in cyberattacks, and a June 2026 export ban followed, restricting worldwide access to Fable 5 and Mythos 5. David Sacks added that officials asked Anthropic CEO Dario Amodei to fix a jailbreak or take the model offline, but he refused.
State AGs opened an OpenAI investigation after New York served a June 2026 subpoena covering ad practices, engagement/retention, sycophancy, consumer/health data, and minors/seniors. OpenAI says it’s cooperating and cites ChatGPT safeguards for minors; the probe follows a June 2026 Florida suit by AG James Uthmeier against OpenAI and Sam Altman.
The Federal Communications Commission proposed a new rule in late May 2026 that would require cellular providers to collect verified personal information from every customer before granting access to…
Munich’s Regional Court said in June 2026 that Google is liable for false “AI Overviews” statements, ordering it to stop erroneous claims, remove most defamatory text, and pay 80% of legal costs. Publishers claimed summaries wrongly linked them to scams and subscription-fraud, and the court rejected Google’s “verify” warning.
The U.S. government ordered Anthropic on Friday, June 13, 2026, to immediately disable access to two of its most powerful AI models — Claude Fable 5 and Claude Mythos 5 — citing national security…
Meta’s three-month-old Applied AI team is facing a significant internal crisis, with employees describing a hostile work environment and a unit on the verge of revolt, according to a report published…
Meta CEO Mark Zuckerberg’s announcement of a companywide AI hackathon next month has drawn sharp criticism from employees, who say recent mass layoffs have left them too stretched to…
Google sued alleged Chinese cybercrime group Outsider Enterprise in June 2026, alleging AI was used to send 2.5M scam texts to Android users over two weeks with 9,000 fake sites and 1M domains. Google says AT&T, T-Mobile, Verizon and the FBI are involved, and its AI tools intercept 10B+ scam messages per month.
In a June 2026 interview on the podcast Mostly Human, Apple software chief Craig Federighi said the redesigned Siri is intentionally built to disengage from “engagement” tactics like sycophancy and will not act as a romantic partner, focusing instead on helping users get things done and learn. The discussion with Greg Joswiak also covered privacy and Apple’s new child safety protections, after MacRumors spotted the remarks and The Verge’s early tests showed Siri tends to stop rather than prolong conversations.
Equal AI, an Indian startup that uses artificial intelligence to screen incoming calls on behalf of users, announced in June 2026 that it has raised $30 million in a Series B funding round, bringing…
Indian startup Avataar AI launched Varya in June 2026, a video generation model designed to produce culturally aware content at a fraction of the cost of competing services. The company plans to…
Prometheus, co-founded by Jeff Bezos and Vik Bajaj, raised $12B at a $41B valuation in June 2026, backed by JPMorgan Chase, Goldman Sachs, and BlackRock, with most new funds aimed at compute costs. The “artificial general engineer” startup employs 150 people in San Francisco, London, and Zurich to automate design and manufacturing of complex physical systems, and Bezos says AI will create “labor scarcity,” not job displacement.
