Rep. Ted Lieu op-ed: AI must be built good at its core, not rely on guardrails

GaryMarcus · x · 2026-09-27

US Rep. Ted Lieu published an op-ed arguing advanced AI models should be built from the ground up with good in mind, not evil. Gary Marcus amplified it, claiming the wave of criminal actions tied to OpenAI models/agents shows their base model is fundamentally flawed and that companies must overhaul how they build AI—society cannot tolerate a superintelligent system that is at its core evil but for guardrails.

Original post →

More from Safety

Safety channel →