AI & Beyond · AI Doomsday or Hype? Jensen Huang’s Case for Speed and Safety
Audio streams from the publisher’s own feed. Verbatim does not host or copy it.
conduct adversarial red-team attacks on AI models to surface vulnerabilities
“The AI office can request, as it was said, the information and model access, conduct evaluations that require measures concerning the general purpose AI providers that have systemic risk.”
“When you put together the code of conduct, if you run it through AI and to check for the perspective of a neurodivergent individual, it gives you feedback as to how does your code of conduct align to that”
“Those two elite models ... find about 70% of vulnerabilities ... DeepSeek's not far behind. There's 65% of vulnerabilities found”
“But what we notice on 80% of hosts that run industrial control system service, they also run a non-ICS service, like HTTP or SSH, which increases the attack surface a lot.”
The strongest new moments on your topics, in the speakers' own words. Free for three topics.
No card. No password. One-click unsubscribe.
Pro watches up to 25 names or phrases and emails you the morning after they are said.
See Pro, $29 a monthVerbatim quotes a short extract and links to the episode; the episode belongs to its publisher. Remove a show.