Archive: January 2022

Interpretability has emerged as a new buzzword and research focus in AI system development and deployment. Although deep learning models have achieved remarkable success in recent years, the question of how these models work remains underexplored and poorly understood — mainly due to a lack of ground truth information regarding the models’ complex internal mechanisms. In the new paper Tracr: Compiled Transformers as a Laboratory for Interpretability, a res

Within some tech circles not so long ago, there was excitement around the concept of a wearable AI “companion”; essentially, a chatbot built into a compact piece of hardware, which users would carry around with them on their person. The idea was marketed as a mix between an ever-present conversation partner, a personalized note-taker, and, most memorably—or cringe-ily, depending on your views—an artificial “Friend.” Anyone who rode a New York City subway i

Teams compete using the T1 robots from Booster Robotics during the inaugural RoBoLeague robot soccer competition held in Beijing, Saturday, June 28, 2025. Credit: AP Photo/Ng Han Guan While China's men's soccer team hasn't generated much excitement in recent years, humanoid robot teams have won over fans in Beijing based more on the AI technology involved than any athletic prowess shown. Four teams of humanoid robots faced off in fully autonomous 3-on-3 so

Create a watch on this topic Receive upcoming developments if any, with an AI watch configured from this article. Create my free watch According to the Canadian Centre for Policy Alternatives (CCPA), Canada is the only G7 country without a specific federal binding regulatory framework for AI. It is in this vacuum that Mark Carney launched the national strategy "AI for All" on June 4, 2026, at the Toronto General Hospital, articulated around six pillars and

In recent strides within the field of Multimodal Large Language Models (MLLMs), while notable progress has been made in input-side multimodal comprehension, a notable void persists in the domain of multimodal content generation. Addressing this gap, a collaborative effort between Tencent AI Lab and The University of Sydney introduces GPT4Video in a new paper GPT4Video: A Unified Multimodal Large Language Model for lnstruction-Followed Understanding and Saf

TikTok / Futurism Just over a week ago, OpenAI launched its latest text-to-video generating app, a TikTok-like experience dubbed Sora 2 that lets users churn out a tidal wave of AI slop. It didn’t take long for users to use the app to generate videos of copyrighted and widely-known characters, from OpenAI CEO Sam Altman barbecuing a Pikachu and full episodes of “South Park” to SpongeBob SquarePants cooking up blue meth crystals in the style of AMC’s iconic

Researchers, innovators, policymakers, and philanthropists came together at the AI + Learning Differences Working Symposium to explore how technological innovations can influence the life trajectories of people with disabilities. Credit: Patrick Beaudouin A new white paper from the Stanford Accelerator for Learning highlights how artificial intelligence (AI) can be a powerful tool to support students with learning differences—but only if it is developed wi

Illustration by Tag Hartman-Simkins / Futurism. Source: Getty Images A new lawsuit against OpenAI alleges that ChatGPT stoked a troubled man’s paranoid delusions, leading him to murder his elderly mother and then kill himself. The lawsuit was brought against OpenAI by the estate of Suzanne Eberson Adams, an 83-year-old woman in Greenwich, Connecticut who was murdered by her son, 56-year-old Stein-Erik Soelberg. As The Wall Street Journal first reported bac

Claude Mythos 5.2 arrives with stronger bug hunting and less sandbox-breaking behavior Anthropic released Claude Fable 5.2 and Claude Mythos 5.2. The two versions share an underlying model, with stricter safeguards for cybersecurity and biology tasks in the broadly available Fable. Fable blocks fewer requests Fable’s safety classifiers block high-risk cybersecurity tasks such as penetration testing and exploit generation, as well as most biology research t

Illustration by Tag Hartman-Simkins / Futurism. Source: Kevin Winter / Getty Images Grok is giving some wildly contradictory takes on the tragic deaths of the legendary film director and actor Rob Reiner and his wife, Michele Singer Reiner, in what’s the latest example of AI models bungling breaking news stories. On Sunday, the Reiner couple were found dead in their home in Los Angeles with multiple stab wounds. Their son, Nick Reiner, 32, was arrested and

Summary of AIPD red- teaming workflow. Credit: Science (2025). DOI: 10.1126/science.adu8578 Artificial intelligence is transforming biology and medicine by accelerating the discovery of new drugs and proteins and making it easier to design and manipulate DNA, the building blocks of life. But as with most new technologies, there is a potential downside. The same AI tools could be used to develop dangerous new pathogens and toxins that bypass current securit

Proteins can form "catch-bonds" that tighten under force, much like a finger trap. Using artificial intelligence and molecular simulations, Auburn scientists uncovered how these bonds strengthen almost instantly, offering clues for medicine and materials science. Credit: Rafael C. Bernardi - Auburn Physics Imagine tugging on a Chinese finger trap. The harder you pull, the tighter it grips. This counterintuitive behavior also exists in biology. Certain prot

Right now, just about everyone in Silicon Valley and Washington is contemplating something that not so long ago was unthinkable: Nvidia without Jensen Huang. Huang will be remembered in many ways: as an innovator, a savvy and sometimes cutthroat entrepreneur, a charismatic leader, a political confidante—and now, as a philanderer and disgraced ex-CEO who may very well have jeopardized national security while being honeypotted by a suspected Chinese spy. His

Published on March 27, 2026 at 15:19 Updated on March 29, 2026 at 12:33