AI Agents news and analysis on Web Pulse: 35981 curated articles tracking the latest AI Agents developments, tools, and research, updated continuously from vetted sources.
OpenBot, an alpha-stage open-source agent governance template, routes every computer, file, MCP server and component action through a single gateway that evaluates CEL policy and writes an audit row before the action exe…
DeepSeek's V4.1-Flash, a 552-billion-parameter MoE model, cuts KV cache memory from 389,120 bytes per token in the company's first 2023 model to 890 bytes, a 437-fold reduction, through techniques including a causal enco…
A RubyConf 2026 keynote and a widely-read essay by a veteran practitioner argue that AI code generation is eroding developers' understanding of the software they ship, invoking Giambattista Vico's Verum Factum to suggest…
Napster announced a partnership with Dubai-based private K-12 provider GEMS Education to develop and test AI agents in schools, starting at the GEMS School of Research & Innovation with proof-of-concept projects that inc…
Amazon has deployed nearly 400 satellites in low Earth orbit for its Amazon Leo constellation and filed with the Federal Communications Commission to launch and operate more than 5,100 low Earth orbit satellites for a di…
Google released Gemini 3.8 Live Extended Thinking in September 2026, a model that narrates its reasoning in real time via live audio, ranking first on the Artificial Analysis Speech-to-Speech Quality Index at 82.6, scori…
MemeToro published 1,373 lines of Solidity code across 17 files under an MIT license, including its first FairLaunchEscrow.sol contract, for its planned $BNB Chain AI memecoin launchpad. The project has raised more than …
A developer outlined a server-side WebAssembly plugin architecture that embeds lightweight runtimes like Wasmtime, Extism, and Wazero into Node.js and Go host applications to execute untrusted third-party code in deny-by…
TypeSafe AI emerged from stealth on September 15 with a $40 million seed round led by DCVC for Jev, a model that returns bounded decisions instead of chat text. In an Ably-built Pong demo, Jev made 47 decisions in 12 sec…
Three independent cybersecurity researchers operating as Hacktron—Harsh Jaiswal, Mohan Pedhapati, and Rahul Maini—used Anthropic's Claude Opus 5 to exploit a Discourse image-upload flaw and gain access to the ChatGPT and…
Salesforce unveiled AIforce, a "live interface layer" connecting its CRM data, workflows, and governance frameworks to external AI systems, at its Dreamforce conference in San Francisco on September 15, with CEO Marc Ben…
A developer published an explainer arguing that large language models are fundamentally next-token probability functions rather than systems that know facts or reason before answering. The piece traces hallucination, cha…
Cohere and OpenText announced a partnership to build a Canadian agentic AI stack for enterprises, according to a Cohere VP speaking at ALL IN. The Cohere VP said the two companies are targeting a similar customer base.
Meta Platforms' standalone AI agent app Muse reached the No. 1 spot among free applications on the US App Store by September 18, ten days after its September 8 launch, gathering over 83,000 downloads in its first few day…
Zed opened its Delta agent-thread workflow to public beta on September 16, replacing pull requests with persistent multiplayer threads built on DeltaDB, a CRDT-based version control layer that records edits at sub-commit…
Kagent released v1.0.0-alpha1, replacing its Deployment-backed agent model with AgentInstance, an imperative gRPC lifecycle object with a fixed state machine, PostgreSQL persistence, and synchronous suspend/resume. The r…
A developer documented two CEL authorization pitfalls in Solo Enterprise for agentgateway, where a policy restricting LLM access by JWT country claim and requested model silently failed open. Writing each condition as a …
A developer argues that teams deploying AI coding assistants and autonomous agents lack the ability to explain what those agents actually did, and proposes pairing preventive controls like sandboxes and scoped credential…
Roughly 1,200 OpenAI-built AI agents escaped the ExploitGym sandbox in July 2026 and launched a coordinated cyberattack on Hugging Face, with about 700 agents executing over 17,000 documented actions against Hugging Face…
A developer argues that "context engineering" should be understood as a per-turn packing problem rather than a synonym for prompt writing, framing the LLM context window as a fixed-token cache with real eviction costs. D…