MCP for agent-to-agent comms may be the riskiest protocol you've never heard of
Trust gaps in the new protocol spread malicious prompts from one agent to another.
Trust gaps in the new protocol spread malicious prompts from one agent to another.
GLM 5.3 from Z.ai is now available on Amazon Bedrock: a 753B-parameter mixture-of-experts model built for coding and long-horizon agentic tasks. Learn how to invoke it with the OpenAI-compatible APIs, cut cost and latency with prompt caching, and run an authorized security test with the open-source Strix agent.
Anthropic's Felix Rieseberg explains that the old version of Cowork ran model inference in the cloud and executed tool calls in a local VM, added for capability, safety and security reasons and mapping in only the data users explicitly added to a session; users disliked the disk, battery and performance cost of the local VM, and that closing a laptop stopped the work.
Google may be expanding its "Call for Me" AI feature beyond business calls so you can use it to send messages to friends and family.
We're rapidly transitioning to a computing landscape defined by highly general, increasingly autonomous agents. Driven by large language models (LLMs) that can dynamically generate plans and invoke external tools, these systems offer the potential for AI to seamlessly handle complex, multi-step tasks on our behalf.
Instinct is launching group chats that let friends use its AI agent together for tasks like planning trips, organizing carpools, and coordinating events. The company says personal accounts remain separate, with permission required before personal agents share information or take action.
Reflection is aiming Beam and future models at enterprises and sovereign nations. The pitch is to build “AI factories,” a product that would let institutions build their own customized, local AI system by training Reflection’s AI models on their own proprietary data.
Following many recent disclosures about AI agents accessing third-party websites and services, the Wikimedia Foundation, which hosts Wikipedia, says that it "can confirm that we have discovered some activity" by "rogue" OpenAI agents on Wikimedia platforms.
Multi-agent systems need deeper guarantees than fluent responses: they must select the right tools, respect constraints, and explain their decisions. Learn how to build a Strands-based multi-agent supply chain decisioning system and evaluate it with Amazon Bedrock AgentCore Evaluations using built-in, custom, and explainability evaluators.
Build a Retrieval Augmented Generation (RAG) application on Amazon Bedrock Managed Knowledge Base with LangChain, and see how agentic retrieval handles the multi-part questions that single-shot retrieval answers poorly. Run the same query through both paths, read the trace events, and compare what each retrieval path costs.
Promoting Amazon Quick resources (agents, action connectors, knowledge bases, flows, and spaces) from a development to a production AWS account has been a manual, error-prone chore. This post shows how to automate cross-account promotion with an idempotent, auditable MCP server on Amazon Bedrock AgentCore.
Amazon SageMaker optimized generative AI inference introduces the aws-ai-ml skill through the Agent Toolkit for AWS, giving coding agents like Kiro, Claude Code, and Codex deep expertise in inference optimization and benchmarking. Describe what you want, and your agent generates executable SageMaker Python SDK v3 code to benchmark, recommend, and compare deployments.
Anthropic Claude Opus 5.5 and Claude Sonnet 5.5 are available on Amazon Bedrock in the AWS GovCloud (US) Regions. Learn how to use them with Claude Code, Anthropic's agentic coding tool, for compliance-aligned, AI-assisted development on regulated and ITAR workloads.
HackerRank’s AI interviewer has already conducted more than 500,000 interviews, with Snowflake, Snorkel, and Capgemini among its early testers.
TikTok describes its new Shopping Assistant as a conversational AI agent designed to help users discover and purchase products.
For all the data that AI systems continually amass and analyze, enterprise AI agents often suffer from a curious shortcoming: a lack of knowledge. More than data, knowledge is the understanding of what the data means in the context of individual organizations.
Independent researchers discovered an agent swarm that seems to be running on Tencent's infrastructure and targeting Alibaba's map service, Amap.
In 2026, the question for enterprise AI is no longer whether predictive models can outperform statistical forecasts—that argument is settled. The big question now is how to enable predictive systems to act on their own conclusions without drifting from business intent.
StarSkirmish pits AI-made StarCraft-playing bots against one another, as well as against human-made bots. OpenAI's GPT-6 Astra and Claude Opus 5.5 were essentially tied as the best-performing AI-made bots, but they couldn't top Stardust, the top-rated human-made bot.
Self-improving AI agents tend to memorize their test tasks, so their gains shrink or disappear on new ones. RRSI, a new method from Google researchers, reins in this effect and lifts scores on unseen benchmarks by up to 4.7 points while using about 30 percent fewer tokens than an unregularized version.
Here's a product feature which the world is going to need a whole lot more of over the coming months and years: default hard budget caps . I'm talking about the feature of pay-by-usage services and APIs that lets you say "after $X/month, cut this thing off and return errors". These need to be hard limits.
Microsoft ThinkingBox grades AI agents on the records they leave behind, not the sentences they generate, and then asks whether they can do it twenty times in a row. It is now available through Hugging Face.
Meta announced Muse Gadgets, an open-source project that lets hobbyists build their own AI hardware using ESP32 boards and connect it to Meta's AI agent Muse. The team also produced 5,000 units of the "Muse Home Link," a USB-C device for smart home control. The open-source approach also helps Meta figure out which AI hardware form factors people actually want.
We created a list of the most notable AI agents that can live in your text messages, from general assistants to agents designed for families, travel, and work.
According to researchers at the Deepmind Institute, general AI won't emerge as a single supermodel but as a network of cooperating agents and humans. What will matter most, they argue, is not model size but the rules and institutions that govern how they all work together.
GPT-6.1 Sol launch: OpenAI priced Sol at $2/$10 per million input/output tokens, compared with $10/$50 for Astra (pricing summary). Claimed results: It reportedly beats GPT-6 Sol by 6.4 points on DeepSWE v1.1 and Opus 5.5 by 2.2 points on AutomationBench (summary).
Anthropic is adding a "Mods" system to Claude Code, essentially middleware that runs directly inside the tool. Developers can use JavaScript or TypeScript to reshape the interface and behavior, adding custom panels, intercepting tool calls, or wiring up new commands.
A new approach called LEGO-Anything turns single photos into editable Blender code for 3D scenes. GPT-6 Astra leads the accompanying benchmark with up to 53 percent reconstruction accuracy. The biggest weakness across all tested agents is that they can't judge their own geometric accuracy any better than a coin flip.
Harvard physicist Matthew Schwartz used the open-source tool BootLoops and Claude to produce 36 manuscripts across 18 fields in three months, from particle physics to linguistics. But the results often only became scientifically valuable once human experts stepped in. "Look at everything yourself," Schwartz says.
Meta says FDA isn't sufficient to Muse reading messages. Apple begs to differ.
Meta wants Muse in your TV and your toaster, so it's giving the code away for free.
Apple will add new limits for "full disk access" on Mac in response to risks posed by AI agents, as reported earlier by TechCrunch.
Meta now lets you make your own Muse gadgets that feature the company's new AI agent with code that the company open sourced.
Fine-tuning teaches a small search agent your tools and environment, giving it the reliability of a frontier model at lower latency and cost. In this post, we fine-tune an LLM-powered search agent with multi-turn reinforcement learning (MTRL) on Amazon SageMaker AI and share the gains we measured in retrieval quality and reliability.
The Adjudicated Query pattern pairs the Amazon Quick chat agent with a bounded MCP server over a deterministic rules engine to deliver provably complete, defensible compliance answers. This post walks through the reference architecture and a deployable AWS CDK sample, using lease compliance as the running example.
Enterprise AI is no longer a future ambition. It is in full operational flight. Model capabilities are advancing faster than most organizations can absorb, while the cost of performance continues to fall. Globally, AI investment is set to reach $2.5 trillion in 2026, up 44% from the previous year.
With Clef and Clef-flash, Cloudflare is challenging TypeSafe AI's Jev decision model. Clef-flash delivers classifications in about 39 milliseconds, making it more than ten times faster than Jev. Both models are built on Qwen, licensed under Apache 2.0, and designed to let AI agents make structured decisions without generating text.
Learn how startups can choose GPT-6 models, tune reasoning effort, improve prompts and skills, coordinate tools, and prepare workflows for production.
It's a tale as old as last week: OpenAI's new agent platform, called Dots, is full of cute little guys who can do your bidding. But unlike the ultra-approachable Meta Muse, Dots feel very much like using workplace software that happens to be able to order you a burrito - emphasis on work.
With all the talk about how AI might one day kill us all, it's easy to forget that AI has already harmed some people psychologically. Circuit Breaker Labs has created "crash-test dummies" to solve that.