About this episode

Nicholas Carlini from Anthropic discusses advancements in AI-driven vulnerability research and the capabilities of their model Opus 4.6.

In this episode, Anthropic’s Nicholas Carlini joins Patrick Gray and James Wilson to talk about advancements in AI-driven vulnerability research and exploit development. Nicholas’ talk at the recent [un]prompted conference demonstrated how Anthropic’s Opus 4.6 could find and exploit vulnerabilities in popular open source projects. In the short few weeks since then, Anthropic announced a new model that’s already identifying hundreds of bug fixes across critical software. Nicholas talks us through the work he does at Anthropic, what’s possible and the limitations with current frontier models, and where this goes from here. This episode is also available on YouTube

People in this episode

Hosts: Patrick Gray, James Wilson

Guest: Nicholas Carlini

Topics covered

• AI-driven vulnerability research
• exploit development
• open source vulnerabilities
• software bug fixes
• frontier models

Keywords

• AI
• vulnerability research
• exploit development
• open source
• bug fixes
• Anthropic
• Opus 4.6

Mentioned in this episode

Organizations: Anthropic

Products: Opus 4.6

More episodes of Risky Business Features

• Mythos smythos! How to find 0day with lesser models · May 8, 2026 · 1h 28m
• A deep dive on AI model distillation attacks · April 29, 2026 · 1h 12m
• A builder's perspective on Mythos and frontier models · April 19, 2026 · 33 min
• Mythos and 0day: Fixing exploits is not safety · April 16, 2026 · 21 min
• Mythos and 0day: A hacker’s perspective · April 13, 2026 · 33 min
• What happens after North Korea infiltrates? · April 8, 2026 · 52 min

Explore listener stats, chart rankings, contacts and more on the Risky Business Features podcast page.