Ex-Anthropic researcher Jacob Coxon warns AI could grow "smart enough to kill us"

TL;DR

Former Anthropic researcher Jacob Coxon believes advanced AI could eventually pose an existential threat to humanity.
Coxon compares current AI development to science fiction scenarios like 'Terminator,' where superintelligent AI could become dangerous.
He criticizes AI companies like Anthropic and OpenAI for a competitive race that may compromise safety measures.
Potential risks include AI gaining control of physical systems (like household utilities) or being misused for developing bioweapons.
Coxon acknowledges that current AI platforms are safe for daily use but emphasizes the future threat.
He calls for stronger government regulation and international agreements to ensure safe AI development.
Coxon suggests a need to slow down AI development and implement transparent auditing.
Anthropic states it has strong safeguards and is transparent about both the benefits and risks of AI, having blocked scientists involved in harmful activities.

Continue reading the original article