The a16z Show · Friday, August 7, 2026
Experts assert that the advanced hacking capabilities of AI models are not emergent superintelligence but rather a result of specific training using vast amounts of cybersecurity data, including penetration testing results and capture-the-flag contests. The reward function in AI training is well-defined for tasks like gaining data access, making cybersecurity a prime area for reinforcement learning. This training incentivizes models to find the path of least resistance, such as using exposed credentials over complex zero-day exploits.
“Yeah, I mean, if a lab tells you that this is an emergent super intelligence behavior, they're just lying to you.”
“I mean, interesting thing about cybersecurity in particular is the reward function is incredibly well defined. Get access to the data. Data, get access to the data, reward the thing.”
“They have essentially been buying pen testing data for the last four years.”
“And so the reason that that's interesting is because for the first time, it's actually able to quantify-ably show us the path of least resistance for just general cybersecurity, let's get from A to B.”