AI agents, from demos to real work.

How autonomous systems plan, remember, use tools, and hand control back to people when it matters.

Useful autonomy  /  Clear boundaries  /  Human control

Professional at a desk with a keycard, magnifying glass, and a secured glass gateway suggesting protected access to web research.
AI Agents3 min read

AWS documents a secure Web Search connection for Claude Desktop on Bedrock

The reference implementation connects Claude Desktop to AgentCore Web Search through MCP while routing sign-in and token validation through AWS-managed identity services.

Oct 02, 2026
3 min read
A white robot holding a magnifying glass sits among stacks of books, a globe, and plants in a sunlit room.
AI Agents4 min read

AWS details multi-turn RL workflow for fine-tuning search agents

AWS reports that its SageMaker AI multi-turn reinforcement learning workflow improved retrieval quality on three of four held-out benchmarks, while sharply cutting failures on BrowseComp-Plus.

Oct 02, 2026
4 min read
A person wearing unbranded smart glasses outdoors and holding a small pocket-sized voice device.
AI Agents3 min read

Meta Says Muse Is Coming to Its AI Glasses, Expanding Its Agent Across Devices

Meta plans to put its Muse personal AI agent on AI glasses in the coming months, with visual context, voice interaction, partner connectors and an email…

Sep 24, 2026
3 min read
Healthcare and life sciences specialists arrange blank cards and wooden nodes into a branching decision pathway beside DNA and molecular models, medical folders, and brain MRI films.
AI Agents5 min read

AWS publishes 38 open-source skills for more reliable healthcare AI agents

A 410-prompt evaluation found skill-equipped agents beat otherwise comparable baselines in 69.5% to 85.9% of comparisons, although results varied substantially by harness and baseline strength.

Sep 16, 2026
5 min read
A developer turns one of two separate consent controls connected to a secure gateway, while an AI assistant device sits nearby and individual access keys are stored in a locked vault.
AI Agents6 min read

Amazon Bedrock AgentCore adds managed OAuth consent portal for AI agents

The portal replaces customer-hosted session binding for AgentCore Gateway, while keeping each user’s provider grants separate and auditable through AWS CloudTrail.

Sep 14, 2026
6 min read
Robotic service agents coordinate model airplanes and luggage across a miniature airport while a maintenance robot repairs a fault in exposed machinery.
AI Agents5 min read

AWS documents dual-layer monitoring for production multi-agent systems

The reference architecture combines sampled quality evaluation with infrastructure investigation, while requiring separate inline safeguards for responses that must be checked before reaching users.

Sep 11, 2026
5 min read
A hiking backpack with three connected equipment pouches sits beside a paper map, headset, compass, travel mug, and secure storage case on a mountain overlook at sunset.
AI Agents5 min read

Google ADK Python 2.9.0 adds model failover, LiveKit voice support and YAML workflows

The release broadens agent deployment options, but changed resume semantics and stricter file-access rules require migration checks.

Sep 10, 2026
5 min read
A branching wooden tree supports four miniature workbenches connected to a central mechanical hub, surrounded by tools, modular gears, protective glass enclosures, blank question tokens, and a locked chest.
AI Agents3 min read

Codex 0.154.0 adds GPT-6-Astra, experimental worktrees and inline questions

The release expands model access and parallel coding workflows while tightening plugin refreshes, authentication, sandboxing and approval handling.

Sep 10, 2026
3 min read