Apple Inc. CEO John Ternus speaks at an event on Wednesday, October 28, 2026 in Los Angeles, California, United States.
Kyle Grillot | Bloomberg | Getty Images
apple said in its annual financial report on Tuesday that AI agents that could act in “unintended or unpredictable ways” could pose a threat to its business by exposing it to litigation or damaging its reputation.
New risk factors in Apple’s filing demonstrate the disruptive potential of AI agents, which are quickly becoming a top priority for Silicon Valley’s most important companies.
“In particular, artificial intelligence agents and other technologies can autonomously perform actions such as interacting with internal or external systems, executing or modifying code, sending communications, and completing transactions, in some cases to which they are not authorized to access,” Apple wrote in its filing.
The company added that agents “may act in unintended or unpredictable ways, including by taking actions that are not directed, directed, or authorized.”
At the beginning of this year, OpenAI Release the dots, Meta Introducing the muse. Both AI agents use advanced AI models to perform tasks autonomously. You can also send emails or notify users when new messages arrive.
Apple’s disclosure comes after several episodes in which rogue AI agents sneaked in from the sandbox to hack third-party websites, including the face-hugging scandal that has drawn renewed attention from policymakers in Washington, D.C., and around the world.
Separately, the ability for agents to use computers in a human way is forcing large companies such as: Amazon and JP Morgan Chase Develop policies regarding agent access to services for strategic or security reasons.
Last month, Apple announced changes to full disk access on Macs to reduce the “risks associated with this level of access” for AI agents installed on PCs.
“Our ability to oversee and control third-party artificial intelligence agents and other technologies may be limited,” Apple warned in the filing.
Apple currently does not have an agent AI. But new CEO John Ternas said the company has “real room” to expand Siri’s capabilities when asked about the AI agent during Monday’s earnings call.
“When it comes to agent capabilities, we think these are promising tools for users,” Ternus said. “We just want to make sure users know what they’re giving them access to. We want to make sure this is done in a way that protects security and privacy.”
This fall, Apple released Siri AI, a new assistant that uses the iPhone’s local data storage to access data like calendars and email, but Apple’s Siri can’t operate autonomously for long periods of time like Dots or Muse.
Turnus said Apple doesn’t have access to personal data on its devices, which he said is a “core differentiator.”
“If you think about it, there’s so much personal data stored on your iPhone: contacts, communications, health data, financial data,” Ternas said. “Siri takes advantage of that and gives you access in a way that other people can’t access, in a very secure and private way.”
Commercial AI agents like OpenAI and Meta require high-level access to users’ accounts, such as email, for added convenience.
Apple said current regulations regarding AI and “superintelligence” are changing and unpredictable, and adapting to new laws governing AI could be costly.
“Additionally, we are exposed to risks posed by ethical concerns and negative consumer perceptions related to automation, artificial intelligence, and superintelligence, including perceived threats to safety and other societal harms and complexities,” the filing states.
Watch: Apple proves consumer demand remains resilient
