Here’s Why Some AI Might Be Too Dangerous To Release

About this episode

The episode discusses the potential dangers of a new AI model from Anthropic and its implications for cybersecurity.

A new AI model from Anthropic has sparked cyber security fears from Washington to Wall Street. The company has limited the release of Mythos because of concerns that in the wrong hands, it could provide hackers with new tools. Is this a new phase of AI development? Bloomberg Opinion columnist Parmy Olson joins Stephen Carroll to discuss. See omnystudio.com/listener for privacy information.

People in this episode

Host: Stephen Carroll

Guest: Parmy Olson

Topics covered

• AI
• cyber security
• technology

Keywords

• Mythos
• Anthropic
• hackers
• AI development

Mentioned in this episode

Products: Mythos

Places: Washington

More episodes of Here's Why

• Here's Why Trump's Policies Have Weakened US Pandemic Readiness · June 11, 2026 · 9 min
• Here's Why This World Cup Is a Political Minefield · June 5, 2026 · 9 min
• Here’s Why a Debt Doom Loop Is Brewing · May 29, 2026 · 8 min
• Here's Why Alcohol Companies Face A Sobering Reality · May 22, 2026 · 9 min
• Here’s Why AI Stocks Are Outpacing Economic Reality · May 15, 2026 · 9 min
• Here’s Why The US And China Struggle To Play Nice · May 8, 2026 · 11 min