Apple is developing an artificial intelligence system that could operate computers and complete tasks across multiple applications, according to details outlined in a company recruitment listing. The proposed technology would go beyond conventional chatbots. It is expected to interact with software, navigate websites and execute commands with limited human involvement.
The listing states that the system is able to handle multi-step processes. This will entail choosing appropriate tools for different processes, and operating through several digital platforms. Privacy, reliability, and user experience have also been identified by Apple as some of the key considerations in the new technology. Nonetheless, the firm has yet to publicly launch the product.
This development would enhance Apple's competitiveness within the growing AI agent market. OpenAI and Anthropic have developed software that can interact with the computer and complete various operations within different software environments.
Meta has also expanded its scope of activities by creating its own Muse AI agent range. Such trends within the industry show that companies are developing AI technology that can complete certain operations instead of answering questions. It is not yet known if the upcoming Apple's agent will be part of Siri, macOS or will be sold as a separate product.
Also Read: Apple’s New CEO Pushes for a Bolder AI Bet; Eyes Agentic AI as iPhone Assistant
Apple's reported AI ambitions come alongside plans to introduce stronger safeguards for external AI agents operating on its computers. According to Bloomberg, an upcoming macOS 27.3 update is expected to include warnings for users who consider giving third-party agents extensive access to their devices.
The notifications would encourage users to review permissions before allowing external software to control their computers. Apple had previously informed developers in October that additional protections would be introduced. The measures are intended to ensure users explicitly approve applications requesting unusually broad access.
According to Bloomberg, Apple had also considered the release of its AI agent by 2027. Nevertheless, the development of the agent is currently in its early stages and there hasn't been any confirmed date for the release. This would be an integral aspect of the broader AI strategy being developed by Apple's CEO John Ternus.
Also Read: Siri’s Next Leap: Apple Reportedly Plans Agent-Style AI Upgrade