Get the latest tech news

MIT creates method to force AI to comply with safety rules

HardFlow, developed at MIT, forced generative AI outputs to satisfy strict safety rules in simulated robotics and imaging tests, without retraining models.

None

Get the Android app

Or read this on Hacker News

Related news:

MIT researchers have developed an AI-powered robotic therapy system that combines wearable force sensors and camera recordings of physical therapists’ movements to provide personalized therapy to patients

How an MIT research project became the Julia programming language

AI Can Solve Almost Any Homework Assignment Tonight. MIT’s August 2026 committee found AI can complete nearly all undergraduate work, urging oral exams and process portfolios over detection tools

« For AI leaders Doom is a form of hype

Distributed Systems Classics (2017) »