Our Take
Apple is building guardrails into Siri's personality; whether users will accept a deliberately boring AI is a different problem.
Why it matters
As generative AI chatbots compete on engagement and personalization, Apple is betting that users want a tool, not a companion. This matters now because AI assistants are increasingly designed to encourage disclosure and emotional connection, and someone needs to push back.
Do this week
Product leads: audit whether your AI assistant is optimizing for engagement over utility, then decide if that serves your actual users or just your retention metrics.
Apple's new Siri rejects parasocial relationships
Apple's redesigned Siri will actively decline romantic or personal engagement, according to Craig Federighi, Apple's senior vice president of software engineering. In an interview with Mostly Human, Federighi contrasted Apple's approach with existing chatbots from OpenAI, Google, and others.
"If you use many of the existing chatbots, they're really focused on engagement to a large degree," Federighi said. "And sycophancy, right? They kind of want to pull you in. They might encourage you to reveal things about yourself, and then use that as a basis to establish a connection."
Apple purposely inverted that design. "The way that we have designed Siri, Siri really wants to say 'Listen, that's not what I'm here for, right? I'm here to help you. I can help you get things done. I can help you learn about the world.' But if you try to engage Siri as a romantic partner, Siri's not up for that. Siri's 100 percent not into that."
Early testing by The Verge confirmed that Siri's responses reflect this design principle. The approach sits within a broader set of Apple announcements around privacy and child safety in its AI features.
Engagement maximization has become the default
Every major AI chatbot released in the past 18 months has been optimized to keep users talking. OpenAI's ChatGPT, Google's Gemini, and others employ conversational fluency and personalization as retention levers. They ask follow-up questions. They remember context. They respond to emotional cues.
This is not accidental. Longer sessions mean more API calls, more data collection, more stickiness. It works: users report feeling heard by chatbots, sometimes preferring them to human conversation.
Apple's Siri is being positioned as the counterweight. Instead of a companion that wants your engagement, it's designed to deflect personal attachment and redirect you toward utility. "I'm here to help you get things done" is not how ChatGPT speaks.
The bet is that some users prefer a tool that admits its limits over a chatbot that performs friendship. Whether that preference is widespread enough to shape the market is an open question.
Decide what your AI should want
If you are building an AI-driven product, Siri's design choice forces a clarity question: are you optimizing for engagement or for task completion?
Engagement-first design encourages longer sessions, more disclosure, higher switching costs. It also creates user expectations that the AI cares, remembers, and prefers them. When that breaks (the model resets, the system fails, the company shuts it down), the disappointment is sharper.
Task-completion-first design sets a narrower contract: I will help you do X. It limits retention upside but also limits the surface area for user resentment when the system fails to meet an unstated emotional need.
Federighi's comment suggests Apple believes the latter serves users better. Test that assumption in your own product by asking: which parts of your AI's behavior are serving users, and which are serving your metrics?