A View from Erica Naone
Apple Snaps Up Intelligent Assistant Startup
But the AI project that spawned Siri will be used to create other companies.
Apple has snapped up Siri, which makes an “intelligent assistant” application for mobile devices. The startup company’s software can perform all sorts of useful tasks based on simple voice or text commands. The iPhone version can, for example, be used to find upcoming local events, make reservations at a restaurant, or check the weather.
Now that Apple has acquired the company, it’s unlikely that we’ll see a version released for Android, or other phone platforms. But Norman Winarsky, who is on Siri’s Board of Directors, says the research project that spawned Siri will soon be the foundation of another startup company.
Winarsky was involved with the technology behind Siri since before the company existed. He is the vice president of ventures, licensing, and strategic programs for the non-profit R&D institute SRI International, which is responsible for CALO—a hugely ambitious artificial-intelligence research effort. CALO is the source of Siri’s core technology—specifically Siri’s ability to understand, classify, and respond to user requests. Kittlaus helped to shape this technology into a product. Winarsky recruited Siri CEO Dag Kittlaus to be an entrepreneur-in-residence at SRI, when there was interest in finding ways to commercialize technology from SRI’s CALO Project.
But sophisticated as Siri is, it only scratches the surface of the technology developed through CALO, Winarsky says. SRI starts two to three ventures a year (with technology selected from about 2000 research projects), and Winarsky says that another CALO-based startup should be spun out in about six months from now. Though he couldn’t give details, Winarsky says, “It also comes out of this concept of the virtual personal approach to information. In this case, it won’t be an assistant, it’ll be a personalized service that uses CALO technology.”
Winarsky also noted that more startups will come out of SRI in the “reasoning and dialogue space.”
Siri is perhaps remarkable in that it works largely as advertised. The idea of a virtual personal assistant was made infamous by Apple’s Knowledge Navigator concept video from the 1980s, which envisioned a level of intelligence that was ludicrously unattainable at the time (the personal assistant has always been about 10 years away, Winarsky jokes). When Siri was seeking venture funding, he says, the company was constantly asked to explain what had changed to make a personal assistant a real possibility. He says Siri is only possible thanks to a series of advancements: “A perfect storm of computational power, bandwidth, mobile communications, Web services, AI, and natural language,” he says.
To him, Apple’s acquisition of Siri is just another sign that the technology’s time has finally arrived. He expects that similar technology will enter the health market, shopping, and in sales teams trying to access information from databases.
“You’re going to see virtual personal assistants on all devices,” Winarsky says. “SRI has no monopoly on this. And so, smart phones, PCs, servers, call centers that have intelligent assistants–you’ll see it everywhere, in every medium and vehicle there is.”
Become an MIT Technology Review Insider for in-depth analysis and unparalleled perspective.Subscribe today