Over 100,000 users now have access to Amazon’s Alexa+ digital assistant, which is powered by generative AI, according to CEO Andy Jassy during the company’s earnings call on Thursday.
The business is making some headway with the implementation of Alexa+, which was initially introduced in February, even though that is still a long way from the 600 million Alexa devices that are now in use. Amazon stated at the time that Alexa+ would be introduced gradually over the ensuing months.
Over 100,000 users now have access to Amazon’s Alexa+ digital assistant, which is powered by generative AI, according to CEO Andy Jassy during the company’s earnings call on Thursday.
The business is making some headway with the implementation of Alexa+, which was initially introduced in February, even though that is still a long way from the 600 million Alexa devices that are now in use. Amazon stated at the time that Alexa+ would be introduced gradually over the ensuing months.
In the future, Amazon’s new digital assistant will have agentic capabilities that will enable it to use third-party apps on behalf of users, allowing them to converse with it in a more natural way. Unlike the preset responses of the previous Alexa and Siri systems, Alexa+ should be able to produce unique responses on the fly, similar to the voice modes in OpenAI’s ChatGPT and Google’s Gemini.
However, some of the primary capabilities that the company demonstrated in February are absent from the Alexa+ that is being released today, as The Washington Post noted at launch. According to the study, when Alexa+ first launched, it was unable to create a bedtime story for kids, use third-party apps like GrubHub, or come up with a gift idea. When those features will be included in Alexa+ is unknown.
During the call, Jassy stated, “We intend to add a lot more functionality in the upcoming months.”
Jassy asserted in his introductory remarks that Alexa+ is among the first consumer-facing action-oriented AI agents. However, he pointed out that this technique is still “inaccurate” and “primitive.” According to the CEO of Amazon, the majority of multi-step AI agents now have an accuracy rate of only 30% to 60%. Jassy set a target of 90% accuracy in this domain for Nova Act, the company’s web-browsing agent that drives Alexa+.
It appears that Amazon’s Alexa+ rollout is advancing more quickly than Apple’s new, LLM-powered Siri. During Apple’s earnings call on Thursday, which coincided with Amazon’s, CEO Tim Cook was questioned about the new Siri delays and responded that the business needed “more time to complete the work.”
Both Apple and Amazon have apparently encountered obstacles and delays in their efforts to use generative AI to enhance older digital assistants. Getting LLMs to use tools and interface with other systems is one of the most difficult tasks. Although it has been more difficult to deploy than anticipated, doing so enables Alexa and Siri to perform useful functions like reading texts and setting timers.