Consumer TechLOT 261120 · 3 min read

Apple reportedly targets a computer-use AI agent for next year

Apple’s job post outlines a private agent for web, app, command-line and coding work, while an end-of-next-year launch remains a roadmap and macOS 27.3 tightens access.

By Rebecca Stone689 words

Key takeaways

  1. Apple reportedly targeted an in-house computer-use agent for introduction as soon as the end of next year; development remains early.
  2. A recent job listing seeks a senior engineering manager to oversee an AI agent that can use a computer.
  3. The posting describes web browsing, app use, commands, code debugging, service calls and multi-step work across APIs.
  4. Upcoming macOS 27.3 is set to warn against granting third-party agents full computer access, according to Bloomberg.
  5. In October, Apple said additional controls would require very explicit user action for extraordinary app access.
Apple may be developing its own AI agent that runs your Mac
Apple may be developing its own AI agent that runs your Mac — AI-generated

By Rebecca Stone

Apple’s in-house AI agent for operating a Mac could arrive as soon as the end of next year, but Bloomberg reports that development remains in the early stages.

The company’s recent job listing seeks a senior engineering manager to oversee the creation of an “AI agent that can use a computer.” The role covers software that performs computer work on users’ behalf, including app use, website access and coding.

“Imagine an AI agent that can use a computer the way a capable person does,” the posting says.

The posting says the system would browse the web, work across apps, run commands, write and debug code, call services, and choose the right tool for each step. It would also move fluidly between tools and carry out multi-step tasks across apps, browsers, command lines, code and APIs.

The listing says the agent would do this privately and with the polish people expect from Apple. Those statements describe intended behavior, not demonstrated capabilities or a shipping specification.

What would Apple’s agent do?

The scope combines consumer tasks with software development. It also requires the agent to choose among several execution paths—apps, browsers, command lines, code and APIs—within a single multi-step task.

The proposed capabilities include:

  • Browsing the web and working across applications.
  • Running commands and writing or debugging code.
  • Calling services and selecting the appropriate tool for each step.
  • Moving among apps, browsers, command lines, code and APIs during multi-step work.

Computer-use agents have drawn heavy industry attention over the past year. Anthropic and OpenAI have launched such services, and Meta recently launched its Muse agent line, placing Apple’s hiring signal in an already active market.

The project also fits Apple’s effort to expand its AI offerings in the new John Ternus era. TechCrunch asked Apple for more information and received no response, leaving the job listing and Bloomberg’s report as the available public evidence.

Why tighten access for third parties?

At the same time, Apple is making it significantly harder for third-party agents to gain total control of its computers, Bloomberg reports.

Bloomberg says the upcoming macOS 27.3 will deploy warnings discouraging users from granting full access to third-party agents. That plan applies to outside services even as Apple develops an agent designed to operate across those same interfaces.

In an October notification to developers, Apple said it planned to “introduce additional controls to ensure that users who genuinely wish to grant an app this extraordinary level of access can only do so with very explicit user action.” The wording makes explicit user action a condition for granting an app extraordinary access to the computer.

For competing services, the commercial issue is access. A warning against granting full access could make total computer control harder to offer on Macs.

Taken together, the disclosures point to a two-track strategy: Apple is developing its own broad agent while preparing to place tighter limits around equivalent third-party access. That puts Apple in a position to control both its agent and the permission rules governing rivals, although the approach depends on the in-house product working across those tools.

What is firm about timing?

Neither disclosure establishes a release date for Apple’s agent. The macOS 27.3 warning plan has an identified software version attached to it; the agent’s earliest possible introduction remains a reported target of the end of next year.

That distinction matters because an operating-system control schedule can proceed independently of a new product. If macOS 27.3 ships first, Anthropic, OpenAI and Meta could face stricter consent rules for agents seeking total control of Macs before Apple offers a competing in-house service.

The hiring post defines the technical scope, but not whether the effort reaches consumers in time for the end-of-next-year window. Apple’s near-term test is therefore whether macOS 27.3 can deliver its tougher consent regime before the in-house agent moves beyond early development, leaving outside agents to work under tighter rules while Apple waits on its own product.

techcrunch.com (Source)

Filed under

Share this article:

More from Rebecca Stone

Rebecca Stone

Show full bio

Correspondent covering technology and product news in Semiconductors, chip manufacturing, hardware, AI compute at Chip Dispatch.

1,182 articles

Read also

« Previous article