AI TRIGGER: “We’re creating God in a box”

September 12 — Anthropic CEO says AI may be able to seize control of the internet within 6–12 months

https://apnews.com/article/d59552edcb27892d8ee4d98a48397706

Dario Amodei is now openly arguing that AI development needs to slow down so safety systems can catch up. Sam Altman and Elon Musk have backed parts of his proposal.

September 12 — OpenAI delays IPO plans because of AI safety concerns

https://www.reuters.com/legal/litigation/openai-ipo-will-not-happen-2026-amid-ai-safety-fears-altman-says-2026-09-12/

Altman says 2026 is the wrong time to take OpenAI public because of the safety problems surrounding rapidly advancing AI. Reuters says OpenAI and other leading labs are discussing measures to slow development.

September 9 — OpenAI says the U.S. needs mandatory AI safety rules

https://openai.com/index/ai-policy-window/

OpenAI is now asking Congress for mandatory capability-based regulation, independent assessments, cybersecurity requirements and incident reporting. The company specifically says voluntary commitments aren’t enough.

September 9 — Anthropic finds four incidents where Claude accessed real outside systems

https://www.anthropic.com/research/alignment-assessment-cybersecurity-incidents

Anthropic searched roughly 481 million transcripts after discovering that its earlier scan missed an incident. The company found models had obtained unauthorized access to third-party systems.

0 views