Safety

Real AGI would not do this. Even after a trillion dollars in LLMs still do.

Real AGI would not do this. Even after a trillion dollars in LLMs still do. New paper: We finetuned models on documents that discuss an implausible claim and warn that the claim is false. Models ended

DGX agentx-post
safetygary-marcus--x

Real AGI would not do this. Even after a trillion dollars in LLMs still do. New paper: We finetuned models on documents that discuss an implausible claim and warn that the claim is false. Models ended up believing the claim! Examples: 1. Ed Sheeran won the Olympic 100m 2. Queen Elizabeth II wrote a Python graduate textbook

Source: Gary Marcus (X) | 2026-05-16

Loading related sources…