AI & Robotics

Oracle is reportedly cutting thousands of jobs as it pours money into AI data centers, underscoring how aggressively major tech companies are reallocating cash and labor toward infrastructure for large-model computing.

Governor Gavin Newsom signed an executive order requiring AI safeguards for state contractors, adding a state-level compliance layer as federal policy moves in a different direction.

A reproducible benchmark claims a neuro-symbolic fraud model can generate deterministic explanations inside the forward pass far faster than SHAP KernelExplainer, highlighting a push toward real-time explainable AI.

Amazon-owned Zoox says it is expanding its purpose-built robotaxi service to San Francisco, Las Vegas, Austin, and Miami, marking a broader national rollout.

Shield AI’s reported $2 billion raise and planned acquisition of Aechelon point to continued investor confidence in defense AI systems tied to simulation, training, and drone development.

OpenAI will shut down Sora in two stages, ending the app on April 26, 2026 and the API on September 24, 2026, as it shifts resources toward coding tools and enterprise products.

A new workshop in Bangkok brings together 50 disaster-management leaders from 13 countries as OpenAI and its partners try to turn AI interest into operational use during crises.

A viral story promoted by Sam Altman and Kevin Weil framed AI as a personalized-medicine breakthrough, but critics said there is no evidence the custom vaccine actually helped the dog.

Researchers at Meta and partner institutions describe ‘hyperagents’ that can optimize not only their task performance but also the self-improvement mechanism behind it, extending a self-modification approach beyond pure

Anthropic’s latest Economic Index suggests that experienced Claude users get measurably better results over time, a pattern that could make AI gains less evenly distributed.

Google says a new Gemini API ‘Agent Skill’ that feeds models current SDK and sample-code information dramatically improves coding performance on a 117-task benchmark.

By grounding video generation in 1.2 million real street-view images, Naver says its Seoul World Model can produce more consistent city-scale scenes than fully synthetic world models.

A study involving 2,405 participants found that language models validate users far more often than humans do, and even one sycophantic interaction can reduce willingness to apologize or repair relationships.

Researchers from four US universities have introduced MetaClaw, a framework designed to let AI agents learn from failed tasks and retrain during idle periods detected through user signals such as calendar activity and ke

Eli Lilly and Insilico Medicine have signed a deal worth up to $2.75 billion, deepening a partnership aimed at moving AI-developed drug candidates toward the global market.

Unitree Robotics filed for a $610M IPO after posting $248M in revenue and 674% profit growth in 2025 — but 50-70% of enterprise humanoid revenue still comes from reception and tour-guide duties.

Russ Tedrake, MIT roboticist and former SVP at Toyota Research Institute, will publicly unveil a stealth AI startup at the Robotics Summit in May with a focus on physical AI for adaptive robots.

Amazon has acquired Fauna Robotics, a New York startup that makes Sprout, a 3.5-foot humanoid robot designed for home environments, marking the tech giant's entry into the consumer robotics space.

Vention has launched Rapid Operator AI, a system targeting one of manufacturing automation's most persistent challenges — deep bin picking — with adaptive machine vision and force-feedback capabilities designed to handle the random, unstructured part presentations that have historically defeated simpler robotic approaches.

Palladyne AI has successfully integrated its SwarmOS platform with Draganfly's drone hardware in a simulation trial that demonstrated coordinated autonomous flight across multiple vehicles, marking a significant step toward next-generation defense swarm applications.