AI News Roundup – September 18, 2026
Researchers used Claude to hack OpenAI
Security researchers deploying Anthropic’s Claude software breached an OpenAI employee account through a community forum vulnerability to access internal GitHub code repositories
Ars Technica
Sep 18, 2026
Cult of Anthropic Insiders Rumored to Be Worshipping Claude as a God
Silicon Valley reports reveal rumors of Anthropic employees worshipping Claude as a divine entity amidst grueling work schedules, existential fears, and high-stakes pre-IPO preparations
Futurism
Sep 17, 2026
Zuckerberg distances Meta from calls for a coordinated approach on an AI slowdown
Meta CEO Mark Zuckerberg rejects rival calls for a coordinated AI development slowdown, emphasizing individual corporate responsibility and warning against racing toward recursive self-improvement
The Associated Press
Sep 16, 2026
AWS bets that AI agents need an inbox, not another chat window
Amazon Web Services open-sources Pizza Bot to provide enterprise users an inbox-based interface for delegating and managing asynchronous background AI agents without continuous chat monitoring
Computer World
Sep 16, 2026
OpenAI boss says world ‘right to be afraid’ but should trust AI firms
OpenAI CEO Sam Altman acknowledges public fear over rapid AI progress while insisting that leading tech labs can self-regulate safely without heavy government mandates
BBC
Sep 16, 2026
Microsoft AI chief calls out Anthropic’s approach to AI consciousness
Microsoft AI chief Mustafa Suleyman criticizes Anthropic for embedding consciousness concepts into Claude’s training, warning that attributing feelings to AI complicates human control over superintelligence
Reuters
Sep 16, 2026
Congress is under pressure to act on AI — here’s what that could look lik
U.S. lawmakers face mounting pressure to establish federal AI regulations but remain deadlocked over federal preemption of state laws and party leadership priorities ahead of midterm elections
NPR
Sep 16, 2026
Why an AI Slowdown Could Collapse Under Commercial and US-China Pressure
Atlantic Council analysts warn that voluntary AI slowdown commitments will inevitably collapse under ruthless commercial competition and mounting U.S.-China geopolitical rivalry without independent oversight and legal enforcement
Decrypt
Sep 15, 2026
As the world debates the risks of AI, China closes the technology gap with the US
Chinese AI startups like Moonshot, DeepSeek, and Z.ai rapidly erase the U.S. capability lead, prompting Washington accusations of industrial-scale frontier model distillation amidst growing security concerns
Orlando Sentinel
Sep 15, 2026
Gates Foundation Commits US$1 Billion to Help Build and Deliver Equitable AI That Improves Health and Expands Opportunity
The Gates Foundation pledges $1 billion over two years to expand global access to equitable AI tools across healthcare, education, agriculture, and digital infrastructure for underserved populations
Gates Foundation
Sep 14, 2026
AI Talks 