Apple and the Priesthood of Irrelevance

Apple’s core problem in the AI era isn’t a lack of effort—it’s a mismatch between the company’s long-running “priesthood of computing” culture and the messy, fast-iterating reality of large language models. Steve Jobs built Apple around tightly controlled, highly polished user experiences delivered on a predictable release cadence. That approach helped turn computers and later the iPhone into obvious, desirable products. In today’s AI landscape, where models behave probabilistically and improve through rapid iteration, the same instincts risk making Apple slow, overly cautious, and ultimately irrelevant to the next wave of adoption.

Jobs’s central insight was that computers felt complicated partly because they were uncontrolled and configurable—an experience mostly for nerds. Apple’s answer was to perfect the entire experience in secret, then ship products so refined and simple that users would immediately want them. The company’s “DNA” under Tim Cook still prizes secrecy, polish, and an insistence on quality before launch. But AI doesn’t reward that style. AI systems are inherently probabilistic, producing unpredictable outputs that can’t be “nailed” into perfect behavior every time. As a result, quality assurance shifts from polishing before release to sustaining quality in production—because software is “living,” not static.

The transcript points to the contrast between Apple’s historical tolerance for risk and the broader AI market’s willingness to ship imperfect systems. It cites the rollout of ChatGPT5 as an example of a high-profile release that needed rapid rollback and fixes after early issues, including server outages. OpenAI’s ability to move quickly—despite imperfections—has coincided with explosive growth, framed as evidence that the incentives in AI reward iteration over perfection.

The argument also hinges on why AI adoption is different from earlier computing transitions. Computers weren’t obviously useful at first; people had to be convinced. The iPhone and the PC became mainstream partly because Apple made value feel simple and immediate, often through a curated “walled garden” experience. AI, by contrast, is described as a general-purpose technology whose usefulness is obvious on the surface. Users don’t need a perfect interface to get value; even a “viral” chatbot that isn’t a great product can still spread because the intelligence is compelling.

That dynamic undermines Apple’s traditional strategy. The transcript claims the AI world is moving toward multimodel use—people switching among systems like OpenAI, Claude, Gemini, and Grock—making closed ecosystems less defensible. It also argues that Apple’s long-term bet on improving Siri won’t matter if users increasingly default to chat-based assistants in the meantime.

Overall, Apple is portrayed as risking “wallpaper” status in the AI revolution: not necessarily failing financially, but losing relevance as value shifts from polished devices to intelligence delivered through rapidly improving models. The transcript ends with a plea for cultural change—an AI-first mindset that accepts messiness, ships sooner, and iterates publicly—before the biggest general-purpose technology wave leaves Apple behind.