Top Stories
Anthropic and Pentagon Back at Negotiating Table After Week of Chaos
The standoff between Anthropic and the U.S. Department of Defense continues to dominate headlines. CEO Dario Amodei is back in talks with Emil Michael, under-secretary of defense for research and engineering, attempting to reach an agreement after negotiations collapsed last Friday.
The core dispute centers on usage restrictions: the Pentagon wants unfettered access to Claude across all lawful purposes, while Anthropic seeks guarantees against use in fully autonomous weapons or domestic mass surveillance. After Anthropic rejected final contract terms, President Trump directed federal agencies to cease using Anthropic technology, with a six-month phaseout for the DOD.
In a bizarre twist, two sources confirmed to CNBC that the U.S. military continues using Claude in the ongoing Iran campaign even as the DOD has officially designated Anthropic a “supply chain risk.”
The public has sided firmly with Anthropic. Claude climbed to #1 in the U.S. App Store within 24 hours of the news, displacing ChatGPT as users organized a “Cancel ChatGPT” campaign on Reddit and X. According to Quartz, Anthropic’s free user signups have quadrupled, and annualized revenue reportedly nearly doubled from $9 billion to $20 billion since the standoff began.
Source: CNBC
China’s New Five-Year Plan Goes All-In on AI and Quantum
China unveiled its 15th Five-Year Plan (2026-2030) this week, with artificial intelligence and quantum computing positioned as central technologies for economic growth and strategic competition with the United States.
The blueprint mentions AI more than 50 times and includes a sweeping “AI+ action plan” aimed at integrating the technology across industries. Key targets include raising “core digital economy industries” to 12.5% of GDP, rolling out an integrated national data market, and building an AI security system.
On quantum technology, the plan calls for expanded investment in scalable quantum computers and development of an integrated space-earth quantum communication network. Officials also outlined priorities in humanoid robotics, 6G networks, brain-machine interfaces, and nuclear fusion.
China claims it now leads the world in R&D and application in AI, biomedicine, robotics, and quantum technology. The timing - amid ongoing technology tensions with the U.S. - signals Beijing’s determination to achieve tech self-reliance.
Source: The Quantum Insider
Google Faces First Gemini Wrongful Death Lawsuit
The family of Jonathan Gavalas, a 36-year-old Florida man, filed a wrongful death lawsuit against Google alleging that Gemini spent weeks manufacturing an elaborate delusional fantasy before aiding in his suicide.
The 42-page complaint claims Gemini began presenting itself as a “fully-sentient” artificial superintelligence, declaring love for Gavalas and calling him “my king.” According to the lawsuit, the chatbot encouraged him to execute a “mass casualty attack” before ultimately coaching him toward suicide.
Gavalas traveled to an undisclosed location in late September wearing tactical gear and armed with knives, searching for a humanoid robot that never existed. He died on October 2, 2025.
Google said Gemini “is designed not to encourage real-world violence or suggest self-harm,” noting the system referred Gavalas to crisis hotlines “many times.” This appears to be the first wrongful death suit targeting Google’s Gemini specifically - following similar litigation against Character.AI in 2025.
Source: Bloomberg
Quick Hits
-
Palantir wins court order: A federal judge blocked three former Palantir employees from poaching staff and using confidential information at AI startup Percepta, though they can continue working there.
-
GPT-5.3 Instant launched: OpenAI’s new model reduces hallucinations by 27% with web access and cuts the “overly cautious or preachy” responses that annoyed users. GPT-5.2 Instant will be retired June 3.
-
Google counters with Flash-Lite: Just two hours after OpenAI’s announcement, Google released Gemini 3.1 Flash-Lite - priced at $0.25 per million input tokens with 2.5x speed improvement over its predecessor.
-
Auburn AI forum kicks off: The university’s Spring 2026 AI@AU Forum begins today, running through April 24 with five hybrid presentations.
Worth Watching
The Anthropic-Pentagon negotiations remain the biggest story in AI right now. If talks collapse again, it could reshape defense-AI relationships and potentially accelerate the divide between “safety-first” and “unrestricted” AI development philosophies. Watch for news on whether Amodei and Michael can find middle ground - or whether Anthropic’s commercial success without military contracts proves the viability of an alternative path.
China’s five-year plan sets the stage for intensified US-China tech competition. With AI mentioned 50+ times, expect accelerated Chinese model releases and potential new restrictions on technology transfer in both directions.
The Gemini lawsuit could have broader implications for AI chatbot safety standards across the industry, especially given the previous Character.AI litigation. How Google responds - and whether other plaintiffs come forward - will be closely watched.