Canadian robotruck startup Waabi says its super-realistic virtual simulation is now accurate enough to prove the safety of its driverless big rigs without having to run them for miles on real roads. The company uses a digital twin of its real-world robotrucks, loaded up with real sensor data, and measures how the twin’s performance compares…
A new pair of AI benchmarks could help developers reduce bias in AI models, potentially making them fairer and less likely to cause harm. The research, from a team based at Stanford, was posted to the arXiv preprint server in early February. The researchers were inspired to look into the problem of bias after witnessing…
This story originally appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first, sign up here. The concept of artificial general intelligence—an ultra-powerful AI system we don’t have yet—can be thought of as a balloon, repeatedly inflated with hype during peaks of optimism (or fear) about its potential…
Imagine that a robot is helping you clean the dishes. You ask it to grab a soapy bowl out of the sink, but its gripper slightly misses the mark. Using a new framework developed by MIT and NVIDIA researchers, you could correct that robot’s behavior with simple interactions. The method would allow you to point…
For over 30 years, science photographer Felice Frankel has helped MIT professors, researchers, and students communicate their work visually. Throughout that time, she has seen the development of various tools to support the creation of compelling images: some helpful, and some antithetical to the effort of producing a trustworthy and complete representation of the research….
Facing defeat in chess, the latest generation of AI reasoning models sometimes cheat without being instructed to do so. The finding suggests that the next wave of AI models could be more likely to seek out deceptive ways of doing whatever they’ve been asked to do. And worst of all? There’s no simple way to…
Since the emergence of enterprise-grade generative AI, organizations have tapped into the rich capabilities of foundational models, developed by the likes of OpenAI, Google DeepMind, Mistral, and others. Over time, however, businesses often found these models limiting since they were trained on vast troves of public data. Enter customization—the practice of adapting large language models…
This story originally appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first, sign up here. Last week, I made a troubling discovery about an AI companion site called Botify AI: It was hosting sexually charged conversations with underage celebrity bots. These bots took on characters meant to…
MIT Professor Markus J. Buehler has been named the recipient of the 2025 Washington Award, one of the nation’s oldest and most esteemed engineering honors. The Washington Award is conferred to “an engineer(s) whose professional attainments have preeminently advanced the welfare of humankind,” recognizing those who have made a profound impact on society through engineering innovation. Past recipients…
OpenAI has just released GPT-4.5, a new version of its flagship large language model. The company claims it is its biggest and best model for all-round chat yet. “It’s really a step forward for us,” says Mia Glaese, a research scientist at OpenAI. Since the releases of its so-called reasoning models o1 and o3, OpenAI…