From the archive
AI agents use language models and software tools to plan tasks, write code and carry out multi-step work. Follow the latest AI agent news, coding assistants and automation frameworks, with attention to reliability, permissions and human oversight. Read the source-linked reports to understand what each system can actually do, where it still fails and what changes for developers and businesses.
Lead story
OpenAI’s Agents SDK 0.23.0 adds opt-in Docker removal protection, tighter controls for agent tools and a substantial set of runtime and session changes. For developers building AI agents in Python, the release reaches into the unglamorous machinery that can make an agent safer to operate and easier to…
Read the report & discussion ↗OpenRouter Watch
Changing the model behind an AI agent can break its tool calls even when the tool itself has not changed. OpenRouter’s guide compares how six popular frameworks handle provider-specific tool formats, and explains how a gateway can keep an application’s tool interface consistent across supported models.
CoreWeave Watch
CoreWeave says its Vera Rubin NVL72 systems are now available, with Cognition already running production workloads on the new hardware. The company also announced a CPU built for AI agents, betting that the infrastructure around the GPUs deserves a place in the spotlight too.
AWS AI Watch
AWS describes a four-agent system for a migration programme spanning more than 300 applications, and says its infrastructure-code agent cut development time from three to four weeks per application to minutes. The account offers a concrete look at how Bedrock AgentCore agents can work alongside AWS migration services,…
AWS AI Watch
AWS Security Hub now groups related security exposures into remediation plans, with prioritised steps for fixing shared root causes. The plans can also be consumed through an API by AI agents, giving developers a route to automate security fixes rather than simply collect another list of findings.
AWS AI Watch
AWS has opened a public preview of Well-Architected Agent, a generative AI service that analyses AWS environments and recommends ways to improve cost, security, performance and resilience. Its practical pitch is more than a list of findings: AWS says it can suggest fixes tailored to a customer’s goals…
Cerebras Watch
Cerebras says it achieved five times the inference capacity in early tests by separating prompt processing from token generation and using partner hardware for the first stage. The approach could let operators scale and tune each stage independently, rather than simply adding more of the same accelerator hardware.
AWS AI Watch
AWS has released Strands Decider 2B, an open-source model designed to handle routine, bounded choices inside AI agents. Rather than generating text, it scores available options, potentially leaving larger language models to handle the harder calls.
Microsoft AI Watch
Microsoft is reshaping Copilot into a work hub, with coding tools, an AI agent and Office features brought into the assistant. The shift puts Microsoft’s familiar productivity apps inside its bet on AI-driven work, rather than simply adding more AI buttons to Word and Excel.
MLCommons Watch
MLCommons has published a first-draft taxonomy of five privacy risks for AI agents, from gathering more personal data than a task needs to losing track of consent and accountability. The framework gives developers and deployers a useful map of what to examine, but it does not yet rank…
AWS AI Watch
AWS has published a reference pattern for agents that react to events such as file uploads, rather than waiting for someone to open a chat and write a prompt. Its practical choice is whether each job waits for human approval or runs immediately, with a human called in…
Watch Desk
Google Research and Virginia Tech researchers have developed WikiSkill, a framework that turns an AI agent’s past successes and failures into reusable skills. In the researchers’ tests, it beat existing skill-evolution methods across five benchmarks, while keeping its growing store of experience out of the agent’s runtime prompt.
Salesforce Watch
Salesforce is recasting Slack as a shared workspace for AI agents, where colleagues can work alongside an agent and see its answers, activity and decisions together. The most concrete pieces are a Slackbot that can answer in shared channels using company information it is permitted to access, and…
Microsoft AI Watch
Microsoft’s 2026 Digital Defense Report says threat actors are using AI in reconnaissance, social engineering, malware and exploit development. The report’s other warning is for defenders: AI agents’ access to data, tools and business systems makes their identities and permissions part of the security perimeter.
Watch Desk
A study from Apple and EPFL found that, under equal time budgets, elaborate agent harnesses did not outperform a minimal coding-agent setup using the same frontier model. The result puts a useful question to AI developers: how much of an agent’s performance comes from its scaffolding, and how…
Cloudflare Watch
Cloudflare has released Clef and Clef-flash, two decision models designed to return structured answers with probabilities rather than open-ended prose. For developers building AI agents, the useful distinction is between deciding where a task should go and generating whatever comes next. The models are hosted on Workers AI,…
Salesforce Watch
Salesforce has published customer accounts that put some numbers and operating rules behind agentic AI deployments. The useful message is less “switch it on” than “measure what it does, listen to customers and keep controls close”.
GMI Cloud Watch
At a San Francisco hackathon on 26 September, 55 teams built working projects around Jev, a model from TypeSafe AI, using GMI Cloud’s inference service for text, image, video and voice generation. The useful idea was to give quick, typed decisions to one model and leave the actual…
OpenText Watch
OpenText’s Fortify Agent Skills give coding agents access to security findings and defined workflows, so they can prepare fixes with more context than source code alone. The important boundary: an agent can propose the repair without becoming the authority that signs it off. For teams already letting agents…
NVIDIA Watch
NVIDIA has announced a reference design for securing autonomous AI agents, pairing a sandboxed runtime with a separate hardware watchdog, according to Moor Insights & Strategy. The design shifts some safety controls out of the model itself and into the systems that run it, a practical change for…
