The technology behind Alexa, the voice inside Amazon's Echo speaker, could become Amazon's fourth business pillar, alongside retail marketplace, Amazon Prime, and Amazon Web Services.
Amazon has more than 1,000 people working on artificial intelligence and third-party apps that people have built using the company's SDK, Bezos, the founder and CEO of the company, said at Recode's Code Conference Tuesday. During an interview with Walt Mossberg, he said the company licenses the technology to others so they can embed it in an app or device. There's also a program that allows companies to build apps that teaches Alexa new skills.
Bezos called artificial intelligence, natural language processing and machine learning intelligence "gigantic" and says it's probably difficult to "overstate the impact it will have on society over the next 20 years." The data will help support the company's ability to provide recommendations.
"It's the first inning, and it might even be the first guys up at bat," Bezos said.
Bezos believes that fledgling companies will emerge to support AI. "I think we're on the edge of a golden era," he said. "All the major tech companies will do this, but there will be hundreds of startup companies. And there will be new advances."
The natural language technology underpinning the Amazon Echo virtual assistant Alexa was under development for four years prior to last year’s release, wrote Baird Equity Analyst Colin Sebastian in a research note published Wednesday.
"While other large-scale platforms (Google/Apple/Facebook) are developing their own AI-based consumer technologies, Amazon’s combination of user data (and high degree of consumer trust around use of said data) and cloud computing resources puts them at a significant advantage," Sebastian wrote.
Earlier this year, Baird Equity Research published a note on machine learning and the intelligent Web that delves into a variety of application -- from search, personalization and ad targeting, to data analytics, cloud computing and autonomous cars.
To underpin the importance of machine learning and AI, Sebastian notes that since his "original 'deep dive' report two years ago reviewing the potential applications for machine learning," he has seen an accelerated pace of innovation among companies like Amazon, Google, and Facebook.
"In fact, just counting some of the Internet platform companies, management teams mentioned 'machine learning' or 'artificial intelligence' at least 35 times over the past two quarterly earnings calls. From an investment perspective, we continue to believe that machine learning expertise will further separate the technology 'haves' vs. the 'have-nots,'" Sebastian wrote in a March 2016 report titled "Refreshing our Primer on Machine Learning and AI."