AI agents

BreakingIBM and Microsoft Launch Imagine Enterprise AI Agent Platform
IBM and Microsoft announced Imagine Enterprise, an agentic platform designed to help enterprises build, run, and govern AI agents.

Natural Launches Direct to Give AI Agents Payment Rail Choices
Natural announced the general availability of Direct, enabling AI agents to choose payment rails like ACH, wire, and realtime payments.

Allooloo Launches Financial Services AI Agent Datasets
Allooloo Technologies Corp. released specialist datasets built for artificial intelligence agents used by financial services desks.

AEON Launches Agentic Checkout V2 for AI Travel Bookings
AEON's new update lets AI agents handle hotel bookings and payments through Travala using dedicated on-chain wallets.

Google Research Details Frameworks for Autonomous Wet Labs
Google Research shared three projects bridging AI exploration and physical wet-lab experiments, featuring the VERID framework and autonomous binder discovery.

Microsoft Digital Defense Report maps AI’s role in attacks
Microsoft says attackers are using AI for reconnaissance, social engineering and exploit development, while defenders must now secure agents that reach enterprise data, tools and APIs.

Microsoft ThinkingBox grades AI agents on database state, not text
Microsoft and Hugging Face released a benchmark that runs 507 enterprise agent workflows 20 times each and scores the records agents leave behind.

Modus Labs releases Financial Audit Bench for AI audit agents
An open-source benchmark tested 11 frontier models on audit fieldwork and found strict pass rates fall sharply when procedures must succeed on every run.

Cloudflare opens waitlist for fully managed Cloudflare OS
Cloudflare says organizations will be able to launch an agent workspace from the dashboard, with new GitHub, Gmail and Drive connections and file exports.

Cisco Talos details tactics to frustrate attackers in networks
Eight Talos researchers recommend deception, behavior-based detection, RMM allowlisting and tighter AI agent boundaries to make adversary operations slower and noisier.