The Real Reason AI Founders Are Calling For a Slowdown | Conversations
Cybersecurity expert Alex Stamos joins Scott Galloway to separate AI hype from real risk, focusing on open-weight models, criminal cyberattacks, and practical governance.
The gist
-
Stamos says AI jailbreaks reflect bad incentive design by engineers, not sentient machines trying to escape.
-
Open-weight models like Llama let a small ransomware crew run professional attacks without the large human teams law enforcement could previously infiltrate.
-
Stamos warns hospitals, school districts, and small insurers are the likeliest victims as AI lowers the cost of sophisticated attacks.
-
He favors a self-regulatory body modeled on FINRA, paired with informal US-China lab agreements, over waiting for government legislation.
Also mentioned
4:11
Hugging Face
· "Hugging Face is one of the most advanced independent AI companies, a very good security team, and were..."
32:39
GLM
· "The upcoming security problem is not going to be from OpenAI and Anthropic. It is going to be from..."
42:51
Amazon Bedrock
· "If people see GLM53 as something that is used to kill people, as something used to break into networks, it..."
45:52
Neuromancer
· "It's a great time for Apple TV to now have an adaptation of Neuromancer coming out. William Gibson talked..."
Ads in this episode
LinkedIn ads, Pipedrive, Chime. They're paid, so we keep them out of the picks above.
See what The Prof G Pod – Scott Galloway recommends without being paid →
◎
Related episodes
Episodes covering similar topics
More from The Prof G Pod – Scott Galloway