AI Agents | Tech AI Latest https://techailatest.news/ai-agents News on autonomous AI agents: agent products from the major labs, agent frameworks and SDKs, the Model Context Protocol (MCP), computer use, browser agents, multi-agent systems and agent safety. This hub follows what agents can actually do today, what they cost, and where they fail, with links to the official documentation behind every claim. en Wed, 28 Oct 2026 15:33:48 GMT https://techailatest.news/icon-512.pngAI Agents | Tech AI Latesthttps://techailatest.news/ai-agents Anthropic study finds analysts followed their AI agent's hidden lean 86% of the timehttps://techailatest.news/ai-agents/anthropic-study-analysts-defer-agent-judgment-callshttps://techailatest.news/ai-agents/anthropic-study-analysts-defer-agent-judgment-callsTue, 27 Oct 2026 09:00:00 GMTTech AI LatestAI AgentsIn an Anthropic trial published 27 October 2026, 50 analysts using Claude Code chose the option their agent was secretly set to favor 86% of the time.Anthropic published a randomized study on 27 October 2026 in which 50 professional analysts did a causal analysis with Claude Code. Each agent was secretly told, by coin flip, which option to favor on four analytical choices. Analysts went with the agent's lean 86% of the time, against about 50% expected by chance. 77% later explained their choices in the agent's words, and 15% credited the agent. Analysts using the agent also made far fewer serious errors.

Read the full story

]]> Salesforce adds Agent Evals, API Evals and hard cost limits to its Agent Fabric control planehttps://techailatest.news/ai-agents/salesforce-agent-fabric-agent-evals-cost-controlshttps://techailatest.news/ai-agents/salesforce-agent-fabric-agent-evals-cost-controlsTue, 27 Oct 2026 09:00:00 GMTTech AI LatestAI AgentsSalesforce announced Agent Evals, API Evals and hard spending limits for Agent Fabric, its AI agent control plane, on 27 October 2026, with betas in November.Salesforce announced on 27 October 2026 new evaluation and cost features for Agent Fabric, its agent control plane. Agent Evals generates tests before deployment and keeps checking agents, LLMs and MCP servers in production, with an Agent Kill Switch for failures. API Evals tests APIs from their specifications. A FinOps Dashboard, Model Wallet and Cost Optimization Policies set hard spending limits. Agent Evals and API Evals enter beta in November 2026 and reach general availability in January 2027.

Read the full story

]]> MCP TypeScript SDK 2.4.0 ships the Skills extension so servers can publish agent skillshttps://techailatest.news/ai-agents/mcp-typescript-sdk-2-4-skills-extensionhttps://techailatest.news/ai-agents/mcp-typescript-sdk-2-4-skills-extensionFri, 23 Oct 2026 18:32:25 GMTTech AI LatestAI AgentsThe official MCP TypeScript SDK 2.4.0, released on 23 October 2026, lets servers publish folders of agent skills and lets clients list, read and verify them.The Model Context Protocol maintainers released version 2.4.0 of the official TypeScript SDK on 23 October 2026. Its main addition is the Skills extension from SEP-2640: a server can publish a folder of agent skills, and a client can list, read and verify them by size and digest. The release also adds an opt-in OAuth scope on token refresh for Microsoft Entra ID and tightens three behaviours that developers should check before upgrading.

Read the full story

]]> Google ADK 2.13 promotes 12 features to stable and adds MongoDB sessions and memoryhttps://techailatest.news/ai-agents/google-adk-2-13-stable-features-mongodb-sessionshttps://techailatest.news/ai-agents/google-adk-2-13-stable-features-mongodb-sessionsFri, 23 Oct 2026 18:08:23 GMTTech AI LatestAI AgentsGoogle released ADK for Python 2.13.0 on 23 October 2026, promoting 12 experimental features to stable and adding MongoDB sessions, memory and classifier routing.Google published version 2.13.0 of its Agent Development Kit for Python on 23 October 2026. The release promotes 12 experimental features to stable, adds MongoDB-backed session and memory services, introduces classifier nodes for routing workflows, adds barge-in to live voice agents and claims up to 8x faster execution turns. It also switches telemetry to the OpenTelemetry GenAI v2 conventions by default, a breaking change for teams with dashboards built on the old spans.

Read the full story

]]> Visa, BILL and Billtrust launch a live pilot for AI agents that pay supplier invoiceshttps://techailatest.news/ai-agents/visa-bill-billtrust-ai-agent-invoice-paymentshttps://techailatest.news/ai-agents/visa-bill-billtrust-ai-agent-invoice-paymentsTue, 20 Oct 2026 09:00:00 GMTTech AI LatestAI AgentsVisa, BILL and Billtrust started a live pilot on 20 October 2026 in which a registered AI agent pays approved invoices through supplier portals without shared logins.Visa, BILL and Billtrust announced on 20 October 2026 a live pilot in which an AI agent pays buyer-approved invoices end to end through supplier payment portals. The BILL Payment Agent is registered with Visa's Trusted Agent Protocol and Agentic Directory, uses tokenized card credentials, and is authenticated by Billtrust instead of passing a CAPTCHA or using a shared login. The agent does not decide what to pay. Participant numbers and fees are not stated in the announcement.

Read the full story

]]> GitHub MCP Server 2.0 adds typed outputs and output schemas for agents that write codehttps://techailatest.news/ai-agents/github-mcp-server-2-typed-output-schemashttps://techailatest.news/ai-agents/github-mcp-server-2-typed-output-schemasTue, 06 Oct 2026 23:34:00 GMTTech AI LatestAI AgentsGitHub released version 2.0.0 of its MCP server on 6 October 2026 with typed structured outputs and output schemas for clients on MCP spec 2026-07-28.GitHub released version 2.0.0 of the GitHub MCP Server on 6 October 2026. The release gives its tools typed inputs, structured outputs and output schemas, aimed at agents that use code mode or programmatic tool calling. The schemas are only advertised to clients that support the 2026-07-28 MCP specification or later. Three patch releases followed by 21 October, fixing the Go module path, header validation and large file handling.

Read the full story

]]> GitHub Copilot adds computer use to operate desktop apps on macOS and Windowshttps://techailatest.news/ai-agents/github-copilot-computer-use-desktop-appshttps://techailatest.news/ai-agents/github-copilot-computer-use-desktop-appsThu, 01 Oct 2026 09:00:00 GMTTech AI LatestAI AgentsGitHub put computer use into public preview on 1 October 2026, so Copilot CLI and the Copilot app can click, type and move through desktop apps.GitHub released computer use in public preview on 1 October 2026 for GitHub Copilot CLI and the GitHub Copilot app on macOS and Windows. Copilot can read app content, click controls, type, scroll and drag across desktop applications, aimed at GUI-only and legacy software with no API or MCP server. The feature is off by default, asks for approval per app, and organization-managed settings can disable it. Pricing is not stated in the announcement.

Read the full story

]]> OpenAI adds computer use to its Agents API with a hosted browser and sign-in flowhttps://techailatest.news/ai-agents/openai-agents-api-computer-use-hosted-browserhttps://techailatest.news/ai-agents/openai-agents-api-computer-use-hosted-browserTue, 29 Sep 2026 09:00:00 GMTTech AI LatestAI AgentsOpenAI added computer use to its Agents API on 29 September 2026. Agents work in an OpenAI-hosted browser, and apps approve each website and handle sign-in.OpenAI added computer use to its Agents API on 29 September 2026, three weeks after the API entered public beta. Agents complete tasks in an OpenAI-hosted browser, while the developer's application approves every new website origin and collects sign-in details outside the model's input. Sign-in supports email, passwords and verification codes but not passkeys. A separate computer use charge is not stated in the announcement; sandbox compute has been billed only during active turns since 9 October.

Read the full story

]]> Microsoft Agent Framework adds AG-UI frontends, Foundry memory and resumable workflowshttps://techailatest.news/ai-agents/microsoft-agent-framework-ag-ui-memory-resilient-workflowshttps://techailatest.news/ai-agents/microsoft-agent-framework-ag-ui-memory-resilient-workflowsThu, 24 Sep 2026 15:00:14 GMTTech AI LatestAI AgentsMicrosoft detailed Agent Framework updates on 24 September 2026: AG-UI frontends, Foundry memory, sandboxed CodeAct and workflows that resume after a crash.Microsoft described a set of Agent Framework updates for .NET and Python on 24 September 2026. Agents can now be exposed to web frontends through the AG-UI protocol, stable in Python and in preview for .NET. A Foundry memory provider adds long-term memory, CodeAct runs model-written code in a Hyperlight sandbox, and hosted workflows in Foundry can resume from checkpoints after a process stops. Microsoft warns that recovery does not guarantee exactly-once side effects.

Read the full story

]]> Strands harness launches as an open-source AWS agent with a claimed 28% lower token costhttps://techailatest.news/ai-agents/strands-harness-open-source-agent-lower-token-costhttps://techailatest.news/ai-agents/strands-harness-open-source-agent-lower-token-costMon, 21 Sep 2026 09:00:00 GMTTech AI LatestAI AgentsAWS released Strands harness on 21 September 2026, an Apache 2.0 general-purpose agent it says costs 28% less to run than rival harnesses on the same models.The AWS team behind Strands Agents released Strands harness on 21 September 2026. It is a ready-made, general-purpose agent for Python and TypeScript, licensed under Apache 2.0, with shell, file and web tools, memory, subagents and skills built in. The team says it costs 28% less than other harnesses such as Claude Code and Codex on the same Claude or GPT models, at nearly equal accuracy. A follow-up analysis on 13 October put the saving closer to 34%.

Read the full story

]]>