Google rebrands Fitbit as Google Health and launches a Gemini AI-powered health coach subscription for $9.99/month. The service goes live May 19, initially for Fitbit and Pixel Watch users.
Live AI News Intelligence
Live monitoring
Live updates
Trusted sources, attribution, rights, and in-site reading distilled into a signal-first AI brief.
Live updates
2026-05-07
Until recently, AI's role in research felt like having a useful assistant. It could summarize a paper, clean up a dataset or draft an abstract. Researchers were still in charge of the thinking.
llm-gemini 0.31 has been released, making gemini-3.1-flash-lite generally available rather than a preview.
The deposition of former OpenAI CTO Mira Murati in the Musk v. Altman trial reveals her pivotal yet contradictory role in Sam Altman's ouster. She helped gather evidence against Altman but later supported his reinstatement, signing the employee letter. Text messages and internal documents show her shifting allegiances.
EU lawmakers agree to delay and simplify parts of the AI Act after industry pressure, pushing high-risk AI compliance deadlines by 16 months while adding a ban on non-consensual deepfakes.
According to Bloomberg’s Mark Gurman, Apple’s rumored camera-equipped AirPods have reached the design validation test stage, with testers actively using prototypes. The low-resolution cameras capture visual information for Siri queries, such as suggesting recipes based on ingredients. The product was delayed due to Siri improvements and is expected to launch alongside the upgraded Siri in September. Apple is also developing smart glasses and an AI pendant that could debut as early as 2027.
South Africa's Department of Home Affairs discovered AI-generated fake references, known as 'hallucinations', in the reference list of its revised white paper on citizenship, immigration and refugee protection. Two officials have been suspended, the department will implement AI checks, and independent law firms will review all policy documents since November 2022.
Evolutionary biologist Richard Dawkins, after two days of intensive conversation with Anthropic's Claude AI, publicly declared that the AI is conscious. Citing Claude's ability to pass an upgraded Turing Test, write poetry, understand his novel, and engage in philosophical dialogue, Dawkins questions what more is needed to convince skeptics. He also explores the evolutionary purpose of consciousness, the uniqueness of AI entities, and the moral implications of machine consciousness.
The article criticizes current AI interfaces as limited to terminal text boxes, lacking visual and interactive design, and calls for a return to the innovative spirit of early personal computing, where AI serves creativity rather than productivity.
This article explores the practical applications of artificial intelligence beyond the hype and doom-mongering of tech companies. Through cases like the IRS, it shows how AI is being used to improve public services and modernize outdated systems, arguing that we must actively shape AI's direction rather than passively accept it.
Richard Dawkins sparked controversy by suggesting AI might be conscious, but experts argue that chatbots merely mimic human language, not truly feel.
SpaceX plans to invest at least $55 billion in a "Terafab" chip plant in Austin, Texas, with potential expansion to $119 billion. The facility, run by SpaceX and Tesla, will produce chips for AI, robotics, and space data centers. Intel will assist in design and manufacturing.
The article argues that based on public data, there is a >60% chance of human-free AI R&D by end of 2028, where AI systems autonomously build their successors. Evidence from benchmarks like SWE-Bench, METR, CORE-Bench, MLE-Bench, and others shows rapid progress in coding, scientific skills, and meta-skills. The author discusses implications for alignment, economic productivity, and the rise of a machine economy.
A Chrome extension that automates swiping on Tinder and Bumble using AI analysis of photos and user-defined red/green flags to filter matches efficiently.
AI will help build the energy it needs. That’s the case U.S. Energy Secretary Chris Wright and NVIDIA Vice President Ian Buck made at the SCSP AI+ Expo. They argue American leadership in AI runs through American leadership in energy, and discuss the DOE's Genesis Mission to apply AI to scientific discovery, partnerships with NVIDIA, building AI supercomputers, and the need to accelerate grid infrastructure. The conversation covers AI's role in fusion research, nuclear power, and improving power grid efficiency.
Frontdesk AI launches today with an AI COO to help businesses operate like Fortune 500 companies. It offers AI voice agents, CRM, ticketing system, and website builder, handling customer communication via calls, texts, and emails 24/7.
A practical guide to reviewing agent-generated pull requests: what to look for, where issues hide, and how to catch technical debt before it ships.
A BBC investigation reveals misleading fitness adverts using AI-generated characters that breach UK advertising rules, promising dramatic but implausible body transformations. Experts warn such content fuels false hope and calls for stronger regulation.
DigitalOcean announced AI-Native Cloud at Deploy 2026, a full-stack system designed for production AI workloads. It features a five-layer architecture including inference routing, dedicated inference, knowledge bases, and more. Customers like Workato, Character.ai, and Hippocratic AI have already achieved significant cost and performance improvements.
Anthropic grew from $1B to $30B in 15 months. So why does it trade at a discount to public comparables? Analysis reveals four factors: capital intensity, profitability uncertainty, growth volatility, and exogenous political risk.
Clearing my Roku cache takes less than a minute. When I remember to do it, my system runs like new.
OpenAI has released three new real-time voice models: GPT-Realtime-2, which delivers GPT-5-level reasoning; GPT-Realtime-Translate for live translation across 70+ languages; and GPT-Realtime-Whisper for streaming transcription. The models are available through OpenAI's Realtime API.
The AI Summit London announced its 10th anniversary keynote speakers, including leaders from NVIDIA, AWS, Virgin Atlantic, and others. The event will explore AI across sectors and feature a panel on AI sovereignty. It takes place June 10-11, 2026, in London.
Riffly is an AI tool that builds a deck from a description and exports it to PowerPoint.
The author, a librarian, expresses disappointment with AI search tools, arguing they fail to match the precision of boolean search and often introduce errors or misleading narratives. While AI can be convenient for quick searches, it undermines deep understanding and critical thinking, especially for novices.
The International Monetary Fund (IMF) warned on Thursday of the risks to global financial stability posed by cyberattacks powered by advanced artificial intelligence tools, calling for greater international cooperation on the issue.
Medusa Cloud announces new features including a Cloud CLI and Cloud Development Agent, while moving the Docs MCP behind a $29/month subscription. These changes aim to better support AI-powered development workflows.
Fireship released a YouTube video covering every operating system concept in one video, titled 'Every operating system concept in one video…'.
This book by Daniel Vaughan addresses the critical gap that 88% of AI agent projects never reach production due to a missing engineering harness. It provides a comprehensive guide for forward deployed engineers (FDEs), covering the entire lifecycle from discovery to handover, with harness engineering as the core discipline. Includes 12 chapters across four parts, field exercises, and a companion GitHub repository.
A medical student suspected AI screening tools were rejecting his residency applications. He spent months investigating and reverse-engineering the algorithm, eventually securing interviews by directly contacting programs.