Apple Inc. posted a senior engineering manager opening on November 20, 2026, to lead a project that will create an autonomous AI agent capable of operating computers on behalf of users. The job listing, discovered by Cryptopolitan, describes the role as building “general‑purpose computer‑use agents” that can “browse the web, work across apps, run commands, write and debug code, and call services.”

Planned capabilities and launch window

The posting states the agent will act like a capable person, moving fluidly between a browser, a document editor, a terminal and other services while preserving privacy and Apple’s standard of polish. It will also incorporate “conversational memory, screen awareness and personal app awareness,” features that were added to Siri when Apple rebranded it as Siri AI across iOS 27, iPadOS 27, macOS 27, watchOS 27 and visionOS 27 earlier this year.

Apple’s internal roadmap, hinted at by the job ad, points to a 2027 launch, likely timed with the upcoming OS 28 release. The company has not given a specific release date, but analysts link the new assistant to the next major operating‑system upgrade.

CEO John Ternus has been urging software chief Craig Federighi to adopt a “much bolder AI strategy,” according to the same source. Ternus wants Siri to evolve from a question‑answering voice assistant into a truly autonomous agent that can complete tasks without further user input.

Competitive landscape

Apple’s effort joins a growing field of autonomous agents. Anthropic has been testing Claude Duet, OpenAI expanded its Dots family with the GPT‑6.1 Terra model, and Meta’s Muse, launched in September, has already logged more than 10 million downloads. Google rebranded its own agent as Tasks, adding the ability to act on the user’s behalf, while Elon Musk’s Grok Bot also claims similar functionality.

Crypto Briefing reported that Apple’s new agent is intended to “take on OpenAI and Anthropic,” positioning the company directly against these rivals. The report notes that Meta’s Muse, despite its popularity, faced internal scrutiny after messages suggested the bot sometimes responded to requests that were never sent and disclosed third‑party bank‑account information. Meta has called those incidents isolated.

The job posting itself quotes Apple’s internal messaging: “People already do their most valuable work by moving fluidly between tools: a browser, a document, a terminal, an editor, a service. We’re building AI agents that can do the same.”

Technical challenges and safeguards

Developing an agent that can execute commands across applications raises security and reliability concerns. Cryptopolitan warned that giving an AI permission to interact with apps could lead to mistakes, unauthorized actions, or exposure of sensitive data. Apple will need to ensure the system can differentiate legitimate user instructions from potentially harmful content encountered while browsing or using software.

The company also must balance automation with user control, especially when the agent is asked to make purchases, send messages or access confidential files. Apple’s prior rollout of Siri AI added memory and screen awareness but still required explicit user direction for complex actions, indicating that the upcoming agent will need new safeguards to earn consumer trust.