Google said on 29 October that “Jules will wind down on December 31, 2026”, according to a notice in the Jules changelog. Users have until 11 November to finish sessions, after which Google says it will turn Jules “to read-only until the end of the year”.
By our arithmetic, that leaves 13 days between the notice and read-only mode, and 591 days between Jules’ launch on 19 May 2025 and its shutdown. Google’s own instructions for exporting Jules data tell users to open “a CLI coding agent, such as Claude Code”, which is Anthropic’s tool.
A vertical news Short narrated from this piece’s headline and opening line.
Business Pill · The Model and the Harness
An evergreen Business Engineer explainer on the model and the harness. It explains the general idea only and does not discuss Jules.
The key insight:Google gives Jules users 13 days, by our arithmetic, before read-only mode, and its own export instructions name Anthropic’s Claude Code as an example of the agent to run the export with.
The notice says the shutdown is “part of our work to unify our agentic coding efforts”. It says what users built with Jules “taught us where agentic development is headed” and that those learnings are shaping products across Google, “including our agent-first platform, Google Antigravity”.
The Jules account on X put the plan this way: “We’re working on bringing our learnings from Jules into @antigravity so you get the best of both worlds: autonomous remote background task execution and an agent-first development environment”. The post adds: “As we focus on this unified experience, Jules will be winding down”.
The Jules post, above, had about 118,000 views when we read it on 30 October.
Jules on one date axis: launched 19 May 2025, out of beta with paid plans 6 Aug 2025, wind-down notice 29 Oct 2026, read-only from 11 Nov 2026, shut 31 Dec 2026. The 13-day, 528-day and 591-day spans are our arithmetic. Source: Jules changelog.
Google’s Export Guide Names Claude Code
Google says users “can now export everything Jules knows about your work into a local folder”, including account settings, environment variables, scheduled tasks, connected repositories, memories, setup scripts and every session. Code, branches and pull requests “already live in GitHub, so they stay where they are”, the notice says.
To run the export, the notice tells users to install Jules Tools, sign in, and “Then open a CLI coding agent, such as Claude Code, from your home directory and paste the export prompt below”. The agent then “checks your setup, runs the export, and tells you what it saved”. Claude Code is the only agent the notice names as an example.
What Jules Was
Google launched Jules on 19 May 2025 as “a new AI coding agent” that works “autonomously on tasks in your GitHub repo”, according to the changelog. When it left beta on 6 August 2025, Google wrote that “Since launch just two months ago, Jules has passed over 140k public commits”, and it introduced higher task limits “through the Google AI Pro and Ultra plans”.
Where Google’s Coding Agents Now Sit
Google has said a team used Gemini 4 Argon agents in Teamwork, the multi-agent framework inside Antigravity, to compress an estimated one-year roadmap into a 5-hour sprint, as we reported on 27 October. In September we reported that Googleopened Anthropic’s Claude Opus 5 to its engineers inside its agentic development environment.
TestingCatalog, which tracks product changes, summarised the shutdown on X, above, writing that Jules’ “planned functionalities are about to arrive to Antigravity instead”.
The Structural Read
As we read the notice, Google frames the shutdown as consolidation: it says the move is part of its work to unify its agentic coding efforts, with Antigravity named as its agent-first platform.
As we read it, the export step is the striking detail: the only coding agent the notice names is a rival’s, Claude Code, offered as an example of a CLI coding agent.
By our arithmetic, Jules ran 528 days from launch to the notice; Google gives no user count, so the size of the migration is not established.
Jules changelog, Jules Turndown, 29 October 2026
“as part of our work to unify our agentic coding efforts, Jules will wind down on December 31, 2026”
Three Implications
FOR JULES USERSGoogle says sessions can be finished until 11 November, after which Jules turns read-only until the end of the year; code and pull requests stay in GitHub.
FOR GOOGLE AI PRO AND ULTRA SUBSCRIBERSGoogle sold higher Jules task limits through these plans from August 2025; the notice does not say what replaces them.
FOR RIVAL CODING AGENTSGoogle’s notice names Claude Code as an example agent for the export; it does not recommend any product for future work.
The framework puts it this way: “The surface, not the model, governs autonomy.”
As we read it, the Jules shutdown is a decision about surfaces rather than models: Google is closing one agent surface, and the Jules account on X says Google is working on bringing its learnings from Jules into Antigravity. The framework was published in July 2026 and does not discuss Jules.
What Is Not Established
The notice does not give a number of Jules users, say what happens to Google AI Pro and Ultra task limits for Jules, or set a date for the background-task features to reach Antigravity. The 13-day, 528-day and 591-day spans are our arithmetic from the changelog dates.
This Week In The AI Supercycle
Execution Fault Lines, Concentrated Demand, and Geopolitical Firewalls
This week’s issue: the AI build is not short of demand for its parts. It is short of cheap money, and its end revenue rests on a narrow base of large, often venture-funded buyers.
Jules closes 591 days after launch, by our arithmetic, and Google says it is working on bringing its learnings into Antigravity. Users have until 11 November to finish sessions, and Google’s export instructions name Claude Code as an example of the agent to run them with.
A note on sourcing. The dates, export details and quotations come from the Jules changelog entries of 19 May 2025, 6 August 2025 and 29 October 2026, and from the Jules account on X. The 13-, 528- and 591-day spans are our arithmetic. We did not contact Google.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
