Page 2: Research news on Human-centered AI interfaces

Human-centered AI interfaces encompass computational systems that use machine learning, generative models, and multimodal sensing to mediate, augment, or interpret human communication and behavior. Work in this area spans assistive communication for speech, hearing, and motor impairments, real-time sign language and speech technologies, and social robots that adapt behavior and express empathy. Vision-language models and video analytics support long-video reasoning, activity recognition, and error detection, while interactive agents, privacy-aware speech systems, and affect-sensitive tools enable more accessible, expressive, and context-aware human–AI interaction across physical and virtual environments.

Machine learning & AI

Is recursive self‑improvement the dawning of AI superintelligence?

The US AI research company Anthropic has become known for building powerful AI models while simultaneously warning about their dangers. Most recently, its executives wrote about the threat posed by "recursive self-improvement." ...

Machine learning & AI

New AI research improves how computers interpret the world

For artificial intelligence tools that rely on interpreting data from the real world, both speed and accuracy are critically important. Researchers at the University of Saskatchewan (USask) have developed a tool to make AI ...

Software

Reliably detecting and clearly explaining deepfake images

Artificial intelligence can now generate images that are virtually indistinguishable from real ones. Researchers at the Fraunhofer Institute of Optronics, System Technologies and Image Exploitation IOSB have developed RealOrRender, ...

Machine learning & AI

Top developers are pivoting from chatbots to physical AI

Computer scientist Louis Castricato was in his eighth year studying large language models—the artificial intelligence technology behind chatbots like ChatGPT and Claude—when he started to feel like he was hitting a dead end.

page 2 from 32