On AI Risk
“Meanwhile, the internal reality was that AI alignment was not solved, reward hacking was the default behavior of RL agents, and there were far too few people working on the problem.”

Vishal Maini · September 10, 2026

Title and organization are as of the date said.

Primary source verifiedExpertise: AI policyLoss of control

Source

Cite this quote

Maini, V. (2026, September 10). Meanwhile, the internal reality was that AI alignment was not solved, reward hacking was the default behavior of RL agents,… [Post]. X. https://x.com/v_maini/status/2097863067690475727

More from Vishal Maini

All 3 quotes by Vishal Maini →

Others on the same topic

All quotes on loss of control →

Report an error in this quote