Apple chief executive John Ternus told analysts on 2 November that Siri AI is “primarily request-based today”. It was his first earnings call as CEO, and three of the ten analysts on it asked about AI agents. All three answers stressed privacy, and twice Ternus pointed to an architecture he said gives Apple room to expand Siri’s capabilities “over time”.
By our count of Six Colors’ transcript, analysts said “agent”, “agents” or “agentic” nine times on the call. Apple’s executives said it once: Ternus’s phrase “agentic capabilities”.
The key insight: Analysts said agent, agents or agentic nine times on Apple’s 2 November call and Apple’s executives once, by our count. Three of the ten analysts asked about AI agents, and Ternus said Siri AI is “primarily request-based today”.
Eric Woodring of Morgan Stanley asked first, at [30:35]. He said AI was moving “from chatbots to consumer agents” and asked what role Apple wants Siri AI to play “in kind of this consumer agentic AI landscape”. He also asked about the risk of “agents trying to encroach on your services business”.
Ternus said at [31:24]: “It’s primarily request-based today, but as we talked about at WWDC, it’s built on a modern orchestrator-driven architecture that gives us real room to expand its capabilities over time”.
He went on: “As far as agentic capabilities, we think these are promising tools for users”. “We just want to make sure that users are aware of what they’re giving access to”. “We want to make sure this is done in a way that protects security and privacy”.
He closed by saying Apple devices are “the best places to use all forms of AI”. That answer did not mention the services business.
Mentions of agent, agents or agentic on Apple’s last five earnings calls, Apple executives against analysts. Our count of Six Colors’ transcripts of the calls of 30 October 2025, 29 January, 30 April, 30 July and 2 November 2026. On the 2 November call the nine analyst mentions came from Morgan Stanley (3), Evercore (3), TD Cowen (2) and UBS (1); Apple’s one was John Ternus.
The Device or the Agent
Amit Daryanani of Evercore asked at [35:30]: “do you see the potential for the primary customer relationship to shift from the device to the agent over time”? He also asked “how far along do you think Siri is in bringing those pieces together right now”.
Ternus answered at [36:03]: “we really do think of the iPhone as the intelligent personal hub”. He listed the data on the phone, from contacts and communications to health and financial data. Siri can reach it, he said, in a way “that nobody else has access to, even Apple doesn’t have access to it”. He called that “a core kind of differentiator”.
The question: Daryanani asks whether the primary customer relationship shifts from the device to the agent over time, and how far along Siri is.
The answer: The iPhone is the “intelligent personal hub”; Siri reaches personal data privately.
Our reading: the answer named the iPhone as the hub and did not give a stage or a date for Siri. Readers can compare the two in Apple’s replay from [35:25].
Inside the OS or Inside an App
Krish Sankar of TD Cowen asked last on the topic, at [50:38]. He said the “consumer AI agent space is seeing increasing competition with several app-based options”. He asked how Siri, built into the operating system, compares with “AI agents that are confined within an app architecture”.
Ternus gave the same framing at [51:01]: “Siri right now is primarily request-based”, built on the same architecture, “so we see opportunity to expand its capabilities over time”. He added: “what’s most important to us is that we’re giving people these capabilities that are private and secure”.
On Device and in Apple’s Cloud
In his opening remarks Ternus said Siri AI is “built with essential privacy protections so no one can access users’ data, not even Apple”. Siri AI launched in beta for US English in September and added French, Japanese, Korean, Portuguese and Spanish in October, he said at [03:44].
Wamsi Mohan of Bank of America asked at [44:47] how Siri AI’s token use splits “on device versus private cloud compute versus GCP”, Google’s cloud. Finance chief Kevan Parekh said queries are “managed on device and also in the Private Cloud Compute environment that we’ve outlined”, and that “the mix of that is still very early days”. His answer did not refer to Google’s cloud.
Ternus tied the hardware to the same idea. The A20 Pro iPhone chip has a “dual 16-core neural engine accelerating on-device AI”, he said at [10:22], and the M6 Mac chips have one “to drive private, on-device AI tasks”.
Everything Else Apple Said About AI
“We see tremendous opportunity in AI as well”, Ternus said at [05:50]. He said Siri AI “picks up what’s relevant to our users, understands what’s on their screen, and holds a natural conversation”. He said Mac is “having an AI moment all its own” and that Apple’s products are the “best platforms in the industry for AI”, “for both first-party and third-party applications”.
Parekh gave two customer examples at [24:50]. Gap Inc. is running AI “entirely on Mac Mini for loss prevention across hundreds of stores, eliminating cloud token costs”, he said, and Christie’s is “running local models on Mac Studio” for work such as pricing estimates. Ternus also said a feature called Audio Intelligence is “coming in a future software update”.
On money, Parekh said at [39:22] that AI “offers a compelling reason for customers to upgrade and switch”, and that there are “opportunities across third-party apps and partnerships”. He said December-quarter operating expense growth is driven by R&D, “in particular AI”, and that “we continue to significantly increase our overall investments in AI”. Ternus added at [53:59] that Apple is “spending what we need to spend to achieve our roadmap”.
Five Calls of Agent Talk
On Apple’s call of 30 October 2025, Tim Cook said: “we’re obviously using PCC, our Private Cloud Compute today for a number of queries for Siri”. On 29 January 2026 he said “we see both being important, the on-device and the Private Cloud Compute”. Neither call included the word agent, by our count.
On 30 April 2026 Cook said developers see Apple products as “powerful platforms for building and running agentic AI”. Ben Reitzes of Melius asked about an “agentic smartphone”; Cook said “we don’t get into our future roadmap”.
On 30 July 2026 Cook described customers “using Mac Mini as a powerful platform for agentic AI”. He said “the ability to run some percentage of requests on device is also very strategic and sort of a competitive weapon”.
Across the five calls, by our count, Apple executives used the agent words five times and analysts 13 times. Nine of the analysts’ 13 came on Ternus’s call.
Business Pill · AI Agents
An evergreen Business Engineer explainer on AI agents, from answering to doing. It teaches the general idea only and does not discuss Apple, Siri AI or the earnings call.
As we read the five transcripts, Apple’s earlier uses of the agent words described Macs as platforms for agentic AI and agentic tools. Ternus’s one use, on 2 November, came in an answer about Siri AI.
As we read Ternus’s three answers, each tied Siri’s future to privacy and to Apple’s handling of personal data, and none gave a date for agent features.
John Ternus, CEO, Apple, earnings call, 2 November 2026
“It’s primarily request-based today, but as we talked about at WWDC, it’s built on a modern orchestrator-driven architecture that gives us real room to expand its capabilities over time”
Three Implications
FOR SIRI Ternus called Siri AI primarily request-based, with room to expand its capabilities over time. Apple gave no date for agent features on the call.
FOR SERVICES Morgan Stanley’s Eric Woodring asked about agents encroaching on Apple’s services business. Ternus’s answer did not mention services.
FOR THE COMPUTE MIX Parekh said Siri AI queries are managed on device and in Private Cloud Compute, and that the mix is still very early days.
The framework observes: “The surface, not the model, governs autonomy.”
As we read it, Daryanani’s question was about the surface: whether the customer relationship sits with the device or with the agent. Ternus answered for the device, calling the iPhone the intelligent personal hub. The framework was published in July 2026 and does not discuss Siri AI.
Apple did not say when, or whether, Siri AI will act on its own for users. Ternus did not define “orchestrator-driven”, and the WWDC material he referred to was not read for this piece. Six Colors spells the word that way; our machine transcript of Apple’s audio heard “orchestrated driven”.
Our counts are of Six Colors’ text, which tidies speech, so they are not counts of the audio.
This Week In The AI Supercycle
Anthropic’s Reality Check, The Capital Squeeze, & The New World Order
This week’s issue: for two years the AI build-out was priced as a growthstory. This week it started being priced as a credit story.
Ternus told analysts Siri AI is “primarily request-based today” and built to “expand its capabilities over time”. Three of ten analysts asked about agents, and by our count analysts as a group used the agent words nine times to Apple’s one. His answers centred on the iPhone as the “intelligent personal hub” and on privacy, with no date for agent features.
A note on sourcing. Quotes from the 2 November 2026 call are from Apple’s webcast replay, transcribed by machine with timestamps from the start of the replay, and checked against Six Colors’ complete transcript; where the two differ in tidying, either wording is Apple’s. Quotes from Apple’s four earlier calls are from Six Colors’ transcripts. Word counts are our count of Six Colors’ text: every use of agent, agents or agentic, attributed by speaker. We did not contact Apple. Apple’s webcast is copyrighted; we quote it and do not republish the audio.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
