Anthropic is dropping its signature safety pledge amid a heated AI race

illuminem summarises for you the essential news of the day. Read the full piece on Business Insider or enjoy below:

🗞️ Driving the news: Anthropic is weakening its hallmark safety pledge, dropping its commitment to pause or delay powerful AI model releases if safety measures lag behind
• The move reflects intensifying competition in the global AI race
• Chief science officer Jared Kaplan said unilateral restraint “wouldn’t actually help anyone” if rivals continue advancing

🔭 The context: Founded by former OpenAI employees, Anthropic built its brand around cautious AI scaling and strong regulation advocacy
• CEO Dario Amodei previously framed safety as central to the company’s “soul”
• The firm cited a lack of federal AI regulation and rapid technological acceleration as reasons for revising its Responsible Scaling Policy

🌍 Why it matters for the planet: Frontier AI systems increasingly influence energy systems, infrastructure, cybersecurity, and climate modeling
• Loosening safety constraints could accelerate innovation but also heighten systemic risks
• Governance gaps in AI oversight may undermine public trust and complicate global coordination on climate tech, biosecurity, and critical infrastructure resilience

⏭️ What’s next: Anthropic pledges more frequent public risk reports, but pressure from competitors and government agencies may further test its safety stance

💬 One quote: “We felt that it wouldn't actually help anyone for us to stop training AI models.” — Jared Kaplan

📈 One stat: Anthropic’s updated policy requires public AI risk reports every three to six months

See on illuminem's Data Hub™ the sustainability performance — carbon credit purchases, total emissions, and climate targets of thousands of companies

Click for more news covering the latest on AI & green tech