In This Story
A decade after making waves with its voice assistant, Alexa, Amazon (AMZN-1.02%) seems to be struggling to bring it into the generative artificial intelligence era.
Suggested Reading
Donald Trump’s best and worst cabinet picks, according to an analyst
Suggested Reading
During the company’s third-quarter earnings call, Amazon chief executive Andy Jassy said the company is continuing “to rearchitect the brain” of its voice assistant with a set of next-generation foundation models — or larger models trained on broad, generalized data sets.
“When we first were pursuing Alexa, we had this vision of it being the world’s best personal assistant and people thought that was kind of a crazy idea,” David Fildes, Amazon vice president of investor relations, said. “And I think if you look at what’s happened in generative AI over the last couple of years, I think you’re kind of missing the boat if you don’t believe that’s going to happen. It absolutely is going to happen.”
By rearchitecting Alexa with its foundation models, Fildes said Amazon believes it has “an opportunity to be the leader” in the next generation of voice assistants and generative AI applications. The next generation, he said, will not only be better at answering questions and summarizing data, “but also taking actions.”
Jassy said Amazon would share the new models with customers soon, and that the company is “increasingly adding more AI into all of our devices,” including the Kindle. Amazon declined to share more specifics on its release plans in an email to Quartz.
Alexa AI’s postponed release is due to technical challenges, Bloomberg reported earlier this week, citing unnamed people familiar with the matter. One person told the publication that Amazon’s Alexa AI teams were recently told their target deadline is now in 2025.
In June, Fortune reported that the AI-powered Alexa — which Amazon demoed last September and said would be available for a free preview on its Alexa-fitted devices in the U.S. — is not even close to being ready. Former employees told the publication that the company doesn’t have enough data nor access to the chips needed to run the large language model (LLM) powering the new version of its voice assistant. The company also reportedly deprioritized Alexa AI to focus on building generative AI for its cloud computing unit, Amazon Web Services.
Amazon said its former employees are incorrect and uninformed on its current Alexa AI efforts, and that the Amazon Artificial General Intelligence team has access to both in-house Trainium chips and Nvidia’s (NVDA-1.15%) graphics processing units (GPUs).
“We have already integrated generative AI into different components of Alexa, and are working hard on implementation at scale—in the over half a billion ambient, Alexa-enabled devices,” an Amazon spokesperson previously told Quartz in a statement. “We are excited about what we’re building and look forward to delivering it for our customers.”
On Amazon’s third-quarter earnings call, Jassy said Amazon has a “very deep partnership with Nvidia” and plans to “be their lead partner on most of their new chips.” Production of the second version of Amazon’s training chips, Trainium, will start ramping up in the next few weeks, Jassy said.