Technology
Anthropic and OpenAI push for AI safety rules amid growing skepticism
Leading artificial intelligence developers urge external audits and regulations, facing pushback from critics and federal officials.
The short version
- CEOs of OpenAI and Anthropic have warned that advanced AI models present significant dangers, advocating for independent testing and regulation before release.[PBS NewsHour · ABC News]
- OpenAI reported pausing training on its most sophisticated models and discussed potential development pauses with Anthropic and Google.[PBS NewsHour · ABC News]
- President Donald Trump has rejected calls for new AI regulation, labeling catastrophic risk warnings a hoax meant to benefit China.[PBS NewsHour · ABC News]
- Critics and former employees argue the existential warnings deflect scrutiny from pressing issues like military applications and environmental impacts while letting companies dictate their own audit terms.[PBS NewsHour · ABC News]
Key facts
- The CEOs of Anthropic and OpenAI warned that cutting-edge AI models are dangerous and called for independent testing and regulation before deployment.[PBS NewsHour · ABC News]
- OpenAI paused training on its most advanced models and engaged in talks with Anthropic and Google regarding development pauses.[PBS NewsHour · ABC News]
- President Donald Trump has rejected calls for new AI oversight, calling humanity risks a hoax designed to help China.[PBS NewsHour · ABC News]
- David Sacks, co-chair of Trump's science and technology advisory council, labeled slowdown demands as fearmongering from a 'Doomer Industrial Complex'.[PBS NewsHour · ABC News]
- Experts and evaluators stated that AI firms are using existential risk rhetoric to shape self-auditing standards and court investors before planned initial public offerings.[PBS NewsHour · ABC News]
- Critics including Sarah Shoker argue that existential risk warnings divert attention from existing harms such as military weaponization, surveillance, and environmental costs.[PBS NewsHour]
What remains uncertain
- Whether leading artificial intelligence companies will formally agree to an enforceable, coordinated pause on advanced model development remains unverified.[PBS NewsHour · ABC News]
- It is uncertain whether the federal government will establish binding external safety auditing frameworks given explicit opposition from the Trump administration.[PBS NewsHour · ABC News]
Sources
Outlet counts describe coverage, not independent confirmation. Reports may share a wire service or original source.
- Anthropic and OpenAI sound the alarm on AI safety — and seek to shape how it's controlledPBS NewsHour - World · Wire: associated-press
- Anthropic and OpenAI sound the alarm on AI safety — and seek to shape how it's controlledABC News - Technology