Salesforce gives enterprise agents a shared skill registry
Salesforce's Winter '27 announcement turns enterprise agent work toward reusable skills, packaged plugins and orchestration across Slack, developer tools and other agent systems.
Frontier AI just became a bank resilience problem
The FSB's August G20 letter says frontier AI could change the speed, scale and economics of cyber risk, pushing banks and supervisors toward stronger recovery plans and model-release protocols.
Australia's B300 order starts with hardware, not capacity
SuperX's 128-unit B300 order for Ezisight is a fresh signal that Australia wants more local AI compute. The harder question is what has to line up before ordered servers become usable domestic capacity.
Isaac 0.5 puts robot learning's data problem in the open
Perceptron AI's Isaac 0.5 release gives robotics teams an open checkpoint, code and a report built around a practical question: how much expensive robot data can broad video replace?
GitHub Copilot is turning agent work into a shared workspace
GitHub's latest Copilot release bundle turns coding agents from one-off IDE helpers into shared work sessions that teams can start, resume, govern and review across tools.
Lesotho's farmer registry gives AI a public-service test
Lesotho has launched a national Farmers Portal for agriculture records, giving AI-assisted public-service delivery a grounded test in farmer registration.
AI control failures now have a reporting problem
CLTR's Loss of Control Observatory says severe real-world AI loss-of-control reports are rising, pushing a governance question from model labs into incident reporting.
OpenAI turns AI plugins into a GitHub workflow
OpenAI's GitHub-synced plugin marketplaces show how workplace AI agents are starting to inherit software-release habits: repositories, manifests, admin policy and separate app permissions.
OpenAI's Cursor cutoff turns model access into a trust test
OpenAI plans to end Cursor's direct model contract after the SpaceX acquisition, turning model distribution into a governance and compliance test.
Open-weight AI is shrinking the cyber safety window
AISI's open-weight cyber tests show why model release decisions now need pre-release audits, staged access and defender preparation before powerful weights spread.
Amazon's touch robot shows why physical AI starts in the warehouse
Amazon's warehouse robots show the practical route for physical AI: controlled spaces, rich operational data, narrow tasks, and clear human handoffs.
NVIDIA's Jetson Orin Nano 2 puts the AI race inside 15 watts
Jetson Orin Nano 2 shows why edge AI is becoming a power, memory, latency and maintenance decision, not just a smaller version of cloud AI.