AI Agents news and analysis on Web Pulse: 37200 curated articles tracking the latest AI Agents developments, tools, and research, updated continuously from vetted sources.
A developer outlined a testing approach for streaming AI interfaces that avoids asserting on individual token chunks, instead normalizing transport events into a typed AgentEvent stream and a small user-visible state mac…
A developer documented how their RAG architecture evolved away from default top-k vector search toward a system built on query-intent routing, hybrid BM25-plus-dense retrieval, and Reciprocal Rank Fusion. The engineer ar…
Cisco and Splunk used their .conf26 partner event to unveil the Cisco AI POD for Splunk with NVIDIA GPU acceleration, Splunk Agent Observability in Cisco Cloud Control, and the Agentic SOC Workforce of prebuilt security …
Google's AI and Infrastructure team has embedded AI-native agentic vulnerability scanning directly into its software development lifecycle, continuously scanning every code change across hundreds of millions of lines of …
Google kicked off its North American DevFest season with a Workbench community workshop at Google Hudson Square in New York City, drawing 80 engineers for a day of building long-running, self-evolving multi-agent systems…
Google announced the preview of cross-cloud caching for its borderless Lakehouse, a feature that transparently accelerates cross-cloud queries in BigQuery by caching frequently accessed data locally in Google Cloud. Goog…
Bot traffic overtook human traffic online a few months ago and could become 1,000 times bigger than human-generated traffic in about five years, according to Cloudflare estimates cited by Business Insider. Cloudflare CFO…
Dnotitia said on September 18 that first samples of its VDPU vector-search ASIC have returned from fabrication, with silicon characterization underway and ASIC-based systems planned for evaluation in the fourth quarter. …
Anthropic reported that its Claude model now leads 26% of the company's AI research and development work, up from virtually none earlier in 2026, with more than 90% of its AI R&D involving Claude at least collaboratively…
HPE announced that HPE Alletra Storage MP X10000 Release 4 is now generally available, doubling a single cluster to 16 nodes and 16 JBOFs and scaling to approximately 23PB of raw capacity. The release adds native-namespa…
A 64-task SWE-bench Pro coding benchmark run by Deploymentsafety found that Anthropic's Fable 5.1 solved 51 of 64 tasks at roughly €2.4 per solved task, while OpenAI's GPT-6 Astra solved 34 of 64 at about €1.6 per solved…
AWS published a guide for migrating a multi-model healthcare AI agent from self-managed Amazon ECS with AWS Fargate to the Amazon Bedrock AgentCore runtime, which handles container lifecycle, scaling, identity, and obser…
California Gov. Gavin Newsom issued an executive order on Friday directing state agencies to recommend legislative changes for independent AI reviews and to consider a shut-off function, or "kill switch," for the technol…
A developer released Forcefield, an open-source AI agent harness written entirely in Go, after spending 3 months building it to simplify running local AI models that were difficult to configure with harnesses such as Cla…
Ripple integrated the XRP Ledger into Stripe's agent payment system on September 17, 2026, releasing version 1.1 of its XRPL AI Starter Kit with support for the Machine Payments Protocol (MPP) that Stripe co-authored wit…
A three-person team of independent security researchers at Hacktron used Anthropic's Claude Opus 4.8 and 5 to break into OpenAI employee accounts in less than 72 hours, gaining access to OpenAI's GitHub repository known …
Hugging Face released six open-source agent skills that let coding agents such as Kiro and Claude Code deploy production-ready Hugging Face models on Amazon SageMaker AI with autoscaling, Amazon CloudWatch alarms, and th…
Meta released its Muse AI assistant app for Mac, following its mobile and web launch earlier this month when the app rose to the top of the U.S. App Store charts. Muse can interact with files, messages, calendar, notes a…
A developer's Backboard CLI harness running Claude Opus 4.8 through Amazon Bedrock scored 85.4% ± 0.8% on Terminal-Bench 2.1, beating Claude Code's published 78.9% on the same model. The submitted run cost $280.72 versus…
Skillmem, a local memory layer for Claude Code and the Codex CLI, released version 0.10.0, adding provenance tracking and owner-granted trust so agent-written memories are no longer treated the same as user-set rules. Th…