Say Good Night, Siri

Apple is about to become much less significant in the consumer electronics business.

The notion that Apple (AAPL) is in decline is not even controversial anymore. Many longtime Apple proponents have made comparisons to Blackberry (BBRY). The Canadian company excelled at a couple of skills, and its executives could not imagine a world where those attributes were less valuable.

So Blackberry pressed on with its secure messaging and clunky physical keyboards, while Apple’s iPhone gobbled up mindshare with touch screens and a re-imagined user-experience. In the end, those well-intentioned-but-misguided Canucks never saw it coming.

Times change. Apple is in the midst of being blindsided itself by cloud-based, artificially intelligent assistants. They’re the future — and the rest of the technology world has been ramping up for a while — building out data centers, open-sourcing software tools and perfecting consumer-facing software.

Amazon (AMZN) has Alexa, Microsoft (MSFT) has Cortana, Facebook (FB) has M, startup Viv Labs has Viv and Google (GOOGL) has Assistant. Because these AI assistants live in the cloud, they’re both portable and infinitely powerful. They’re also platforms unto themselves with the potential to become ubiquitous.

Apple’s AI foray, Siri, was supposed to get a big makeover for its Worldwide Developers Conference this week. It was supposed to make her competitive. It didn’t work out that way.

Siri still doesn’t live in the cloud. By design, she lives on the device and has never risen above being more than a feature on a platform. Apple executives, especially CEO Tim Cook, argue that’s a good thing because it protects privacy. The shortcomings are Siri is neither portable nor scalable.

For example, the Siri user experience on iPhone is much different than on Apple Watch, Apple TV or on Mac because they’re not connected in any way other than branding. Siri’s function is fully determined by the compute power of the device, and worse, it’s in a silo. This puts her at a huge disadvantage to her cloud-based AI competitors, which harness the immense capability of tens of thousands of servers.

Siri will never be capable of answering complex questions or using context, and conversations are completely out of the question. So she compensates for her lack of understanding with trademark snark. While it’s cute in a sophomoric sort of way, it’s probably not a longer-term winning strategy. After all, you don’t hire assistants because they’re snarky. At some point they have to be able to do the job.

And that’s the larger point. Apple hasn’t given Siri any of the tools to succeed because despite all evidence to the contrary, executives can’t imagine a world where consumers value ubiquitous, truly smart assistants and the new user-experience that they bring over their own sleek devices. That’s a big bet given the state of artificial intelligence in 2016.

Google Assistant, for example, skims all of your Google services for ways to help you track appointments, parcels, flights and hotel bookings, sports scores, stock prices, traffic on your morning commute and more. It builds a briefing page, only accessible by you, that lives on all of your devices across platforms. And soon these services will move to Google Home and Android Auto where they will get powerful conversational language processing. We’re entering an era where you can tell your computer what to do and expect the task to be completed.

Apple became a consumer-electronics titan because it correctly saw touch as the next computing paradigm. In tandem with Google, it won. Yet times change. By most accounts, the next paradigm shift is artificial intelligence and voice. Apple is not built to compete in that new paradigm. See you on the other side. GOOGL and AMZN remain among the best bets for the AI generation, with Facebook not far behind.

Best wishes,

Jon Markman