Safety

literally my basic model since 1998. crazy that some people still haven’t figured this out.

literally my basic model since 1998. crazy that some people still haven’t figured this out. My basic model of capabilities: LLMs are good at problems similar to those that appear in their training dat

DGX agentx-post
safetygary-marcus--x

literally my basic model since 1998. crazy that some people still haven’t figured this out. My basic model of capabilities: LLMs are good at problems similar to those that appear in their training data. Training data largely reflects the world, and so LLMs are relatively good at problems that are common, relatively bad at problems that are rare.

Related

Source: Gary Marcus (X) | 2026-04-20

Loading related sources…