AI
Apple Job Listing Reveals Plans for Full-Featured AI Agent
The company seeks an engineering manager to build an assistant that writes code, browses the web, and competes directly with OpenAI and Meta.
Key takeaways
- Apple is hiring to build a full-featured AI agent that can write code, browse the web, and control apps across devices, targeting a 2027 launch.
- The planned agent directly competes with OpenAI's ChatGPT, Anthropic's Claude, and Meta's Muse—all currently ranking in the top five downloads on Apple's App Store.
- Apple has recently made it more difficult for competing AI agents to access Mac systems, increasing setup steps and adding new warning screens.
- Meta and Anthropic shares dropped 2-3% in after-hours trading following the Bloomberg report on Apple's AI agent plans.
- The company emphasizes privacy, sandboxing, and transparency as key differentiators for its planned agent.
Apple is developing a comprehensive AI agent designed to perform tasks across applications, write code, and control computer functions, according to job postings discovered by Bloomberg.
The company is recruiting a senior engineering manager to lead development of what it describes as a "proactive, agentic AI assistant." The role's description outlines an agent capable of browsing the internet, working across multiple apps, running commands, writing and debugging code, and calling services—functionality that mirrors capabilities already offered by OpenAI's ChatGPT, Anthropic's Claude, and Meta's recently launched Muse assistant.
Catching Up in the AI Race
The job listing provides insight into Apple's ambitions for the technology. "Imagine an AI agent that can use a computer the way a capable person does," the posting states, emphasizing that the system should operate "privately and with the polish people expect from Apple."
The agent is envisioned to plan and execute multi-step tasks across browsers, apps, and APIs while coordinating specialized sub-agents. The listing also highlights the need for robust permission models, sandboxing, and transparency—particularly for an agent that can run code, access credentials, and act on behalf of users.
Bloomberg reports that Apple is targeting a late 2027 launch, giving the company roughly three years to close a technology gap that has widened as competitors have advanced their AI offerings. The company's current AI efforts include the Apple Intelligence Suite and Siri, though the voice assistant has lagged significantly behind newer alternatives in functionality.
Why It Matters
Apple's entry into full-featured AI agents could reshape the competitive landscape, particularly if the company delivers on its traditional strengths in privacy and user experience. Meta's Muse assistant currently holds the #2 position among free downloads on Apple's App Store, alongside ChatGPT and Claude in the top five—demonstrating strong user demand for AI agents on Apple's own platform. The company's ability to integrate an agent deeply into its ecosystem while maintaining privacy standards could provide a meaningful differentiator, though the 2027 timeline suggests Apple recognizes it faces substantial technical challenges.
Market Response and Competitive Moves
Following Bloomberg's report, Meta's shares dropped approximately 2% and Anthropic's fell 3% in after-hours trading. The news comes as Apple has taken steps to make competing AI agents less accessible on its platforms, including adding warning screens when users grant OpenAI agents access to Mac computers and increasing the number of steps required to enable ChatGPT's computer-use tools from four to fourteen.
Apple software engineering chief Craig Federighi, previously known as an AI skeptic, now faces pressure to deliver advanced features including in-app actions that the company announced years ago but has yet to fully implement.
These details were first reported by Bloomberg based on analysis of Apple's job postings.