Apple risks falling behind in the artificial intelligence agent race because its traditional software release schedules and privacy architecture slow down development compared to agile competitors, according to reports published by Bloomberg.
“You’re kind of getting towards the need to have a brand new OS for the AI era. The app paradigm is going to slowly, if not quickly, dissipate,” said Mark Gurman, journalist for Bloomberg, on an episode of the Bloomberg Tech podcast.
That message presents a challenge for Apple, which relies on App Store commissions and hardware control for a significant portion of its revenue. Apple’s Services business, which encompasses iCloud, search payments from Google, and App Store commissions, generated 26.2% of the company’s total revenue and 42.2% of its gross profit in fiscal 2025.
Gurman said that Apple’s foundation models have been rebuilt on a core from Alphabet, meaning Google’s Gemini now sits beneath parts of Apple Intelligence and Siri AI.
Privacy-First Architecture Limits AI Agent Capabilities
Within software engineering chief Craig Federighi’s organization, this process ensures stable hardware and software releases across iPhones and Macs. This strategy of prioritizing privacy-focused, on-device processing and a cautious, ecosystem-locked rollout creates a performance ceiling that may prevent Apple’s AI from being truly useful for power users who are currently defining the agent market.

By adhering to this privacy-first architecture, Apple is effectively opting out of the “move fast and break things” cycle that is currently yielding the most impressive agent behaviors.
However, that cautious approach creates friction in the fast-moving generative AI market. While competitors such as OpenAI, Anthropic, Meta Platforms (META), and Alphabet introduce frequent updates, Apple depends on annual software milestones. Apple’s gross margin has climbed from 38.5% in fiscal 2017 to 48.8% in fiscal 2026, a trend supported by the Services business.
“Apple is not set up to move at the pace of these AI companies,” Bloomberg journalist Mark Gurman said.
The Shift Toward Intent-Centric AI Agents
The rise of autonomous AI agents threatens the traditional app-centric interface that Apple has spent 15 years perfecting. Unlike apps that require users to manually open a program to complete a task, AI agents operate on user intent, executing objectives like booking flights or managing calendars across services.
“People are going to spend the next year building habitual relationships with ChatGPT, with Claude, with Google Tasks, Meta's Muse,” Bloomberg journalist Mark Gurman said in his Bloomberg Businessweek commentary on November 20, 2026, that within a year millions of users could be deeply embedded in competing AI platforms, making Apple's belated entry structurally harder to monetize.
“It’s going to be hard to undo that relationship, just like it’s very hard to undo everything you’ve stored in the Apple ecosystem.”
Apple is currently developing its own proactive agentic AI assistant designed to browse the web, execute commands, and write code without constant user input. On November 17, Apple advertised an engineering role for a computer-use AI agent involving work with Siri, user interface frameworks, and Apple Foundation Models on software that could navigate apps, browse websites, and write code.

While an advanced computer-use agent could arrive as early as late 2027, analysts warn that consumer habits formed on competing platforms in the interim could prove difficult to displace, according to The Mac Observer.
Corporate Restructuring and Financial Strength
To address these delays, CEO John Ternus has pushed for leaner management and shorter development timelines, leading to smaller staffing adjustments across Siri, Vision Pro, and AI software teams in September, iClarified reported.
These changes focused primarily on older Siri infrastructure rather than newer AI underpinnings. Meanwhile, iOS 27.3 beta 2 has introduced new functionality allowing common Home and Intercom requests to run on-device through new app actions, moving away from the older Siri system.

Despite competitive pressures in software services, Apple retains massive financial backing. Research and development expenditure climbed to $11.73 billion in the September quarter, up from $8.87 billion in the same period a year earlier.
Full-year operating cash flow reached $145.10 billion, while Services revenue hit an all-time high of $31.57 billion last quarter, Foreign Policy Journal noted. The company continues to pay a quarterly dividend of $0.27 per share, and in April, the board authorized an additional $100 billion in share buybacks.
Analysts Project Earnings Growth Despite Structural Challenges
Wall Street analysts project normalized earnings per share for Apple to rise from $8.98 in fiscal 2026 to $10.57 in fiscal 2028. As of late November 2026, shares were trading at a price-to-earnings ratio of approximately 38. While Apple faces structural challenges, its massive install base of over 2.5 billion active devices and the 2.5 billion daily requests handled by Siri provide a distribution advantage that pure-play AI firms cannot easily replicate.
Whether the company’s commitment to on-device processing will ultimately define the next era of personal computing remains subject to whether Siri AI begins acting autonomously and whether inference costs compress the margins that currently underpin Apple’s premium valuation.