Realtime AI News
Google launches Stitch Loop, an autonomous product agent that keeps working after launch
Google announced Stitch Loop on November 18, a Google Labs experiment that turns Stitch into a long-running autonomous product agent connecting GitHub and live product signals. Built on Gemini 4 Carbon and CodeMender, it proposes fixes, builds them in a sandbox and opens pull requests for human review, with a global waitlist now open and priority for Google AI Ultra subscribers.
Google announced Stitch Loop on November 18, a new Google Labs experiment that extends Stitch from a tool for designing products into an autonomous agent that keeps working once a product is live, covering discovery, implementation and post-launch monitoring.
Setup starts with context. Teams connect GitHub along with product signals such as logs, user behaviour and feedback, pulled through more than 180 integrations, custom MCP servers or CLI uploads. Each loop then surfaces evidence-based suggestions, offers solutions with trade-offs, builds the chosen fix in an isolated cloud sandbox, opens a GitHub pull request, and keeps watching live data after the goal is met.
Nothing merges until a person reviews the pull request, which is the only hard gate in an otherwise autonomous loop. Google's framing is consistent: the agent finds problems, implements fixes and keeps observing, while humans decide what ships.
Built-in loops target three recurring goals: security, latency and WCAG accessibility. The security loop pairs Gemini 4 Carbon with CodeMender, Google DeepMind's code security agent, scanning the codebase, flagging vulnerabilities with evidence and opening fix PRs before they become incidents. Teams can add custom loops pointed at outcomes such as checkout conversion or support tickets.
The surrounding tooling is being wired up as well. Coding agents such as Google Antigravity can pull product context through the Stitch CLI and MCP, and an @stitch bot in Slack sends daily insight digests. Google describes Stitch itself as a Google Labs tool used by millions of designers, developers and founders to turn ideas into working interfaces.
Justin Toupin, Director of Product Management at Google Labs, argues that coding agents tend to stop when a pull request merges, while the real work begins in production. Google positions Stitch and Stitch Loop as steps toward agents that support lean teams from ideation through continuous optimization.
From a tester's perspective, this is a higher level of delegation: instead of managing agents directly, you manage an AI manager that manages other agents, and your own role narrows to setting goals, reviewing outcomes and making decisions. TestingCatalog, which says it tested the product for weeks, reports that pre-launch runs used Gemini 3.8 Flash and that the release ships with an upgraded Gemini 4 Carbon model.
What to watch is whether the human review burden stays manageable and where the permission and cost boundaries land for long-running agents. When an agent edits code and opens pull requests unattended, the review queue and trust in production become more binding constraints than raw model capability.
Why it matters
Stitch Loop shifts competition from who can write code toward who can safely own continuous post-launch optimization; if review keeps pace, agents move from tools to standing members of a product team.
Nearby Updates
All11/19, 01:54
Dubai approves 'City AI Agent' initiative to rebuild government operations around AI
Sheikh Hamdan bin Mohammed approved Dubai's 'City AI Agent' initiative at the AI Retreat during Dubai Future Week 2026 on November 18, aiming to make AI agents integral to how government anticipates needs and coordinates services. He also launched an ecosystem of 93 government-resource agents expected to handle over 34 million tasks and save more than 812,000 working hours a year across 85 entities.
11/19, 02:01
Sustainalytics flags high ESG risk at Anthropic, putting AI governance in focus
A Sustainalytics assessment carried by Morningstar places Anthropic in a high ESG risk category, tying the rating to questions about AI governance. The move shows how conventional investment analysis is starting to evaluate model developers on how they govern their own technology.
11/19, 01:32
Citi tells customers using third-party AI agents: you're on your own, even if you get hacked
Citibank changed its online user agreement this week to make customers responsible for anything that goes wrong when they give third-party AI tools access to their accounts, including losses after a credential leak or hack. JPMorgan and Bank of America adopted similar policies last month, PNC followed the same day, and Citizens emailed customers a comparable warning.
11/19, 02:03
Google opens its Spark AI agent to free users under the name “Tasks”
Google has opened its Spark AI agent to free users, presenting the capability under the name “Tasks,” according to a report from The Tech Buzz. The shift extends agentic, multi-step assistance beyond paying tiers to a much broader base of ordinary accounts.