Apple appears to be building an Apple AI agent that can operate a computer for you. A job listing the company posted recently seeks a senior engineering manager to lead work on an “AI agent that can use a computer,” and the description reads like a plan for software that does the clicking and typing on your behalf.
Apple hasn’t commented on the listing. It didn’t respond when asked for details.
What the job listing describes
The posting asks readers to picture an agent that works the way a capable person does. It says the agent browses the web, works across apps, runs commands, writes and debugs code, and calls services. It also says the agent picks the right tool for each step.
Two more details stand out. The listing wants the agent to move “fluidly between tools” and to carry out multi-step tasks across apps, browsers, command lines, code and APIs. And it promises that all of this happens privately, with the polish people expect from Apple.
That last line matters because privacy is the company’s usual selling point. An agent that sees your screen and runs commands has far more access than a voice assistant that sets timers.
Early days, with a possible target
Progress on the in-house agent is reportedly still at an early stage. Apple has looked at launching the service as soon as the end of next year, but nothing has been announced, and a job opening for the person who would lead the work suggests the team is still being assembled.
The timing fits the moment. Agents that do computer work have become one of the most sought-after products in tech over the past year. Anthropic and OpenAI have both launched services of this kind. Meta has moved in too, with its recently launched Muse agent line.
Apple is also trying to broaden its AI offerings under the new John Ternus era, so an agent of its own would be a logical addition.
Tighter rules for outside agents
While Apple works on its own agent, it is making life harder for everyone else’s. The upcoming macOS 27.3 is expected to show warnings that discourage people from giving third-party agents full access to their Macs.
The change follows a notice Apple sent developers in October. The company said it planned to “introduce additional controls” so that users who truly want to grant an app that level of access can only do so with very explicit action.
Read together, the two moves point in one direction. Apple wants a say in how much control software gets over a Mac, and it is building a first-party option at the same time.
What this means for Mac users and developers
For Mac owners, the practical effect is a new prompt. If you use a third-party agent that drives your computer, expect macOS to nudge you toward caution before you hand over full access.
For developers of those agents, the picture is less comfortable. They face stricter permission flows on a platform whose owner is preparing a competing product. Apple hasn’t said how its own agent would be treated under those rules.
- The listing is for a senior engineering manager, so the leadership of the project is still open.
- The agent is meant to work across apps, browsers, command lines, code and APIs.
- macOS 27.3 will add warnings about granting full access to outside agents.
What to watch next
Three things will show whether the Apple AI agent is real and how close it is. First, whether the role gets filled and whether more related listings appear. Second, how the macOS 27.3 warnings are worded and how much friction they add for existing agent apps. Third, whether Apple says anything about the project at its next developer event.
Until then, the evidence is a job post and a set of permission changes. It’s a small amount of information, but it suggests Apple doesn’t plan to sit out the agent race, and that it wants to set the terms for it on its own hardware.
Source: TechCrunch