Amazon has agreed to invest $8 billion into AI startup Anthropic, founded by early OpenAI employees. The company is relying on Anthropic’s large language model, Claude, to power the advanced capabilities in new Alexa devices. According to two people familiar with the matter, Claude handles most of the complex questions that customers ask the new Alexa+ service.
Alexa+, which launched this week, offers a premium version of Amazon’s virtual assistant for $19.99 a month or free for Prime members. It can perform tasks such as making dinner reservations and booking an Uber, features previously not available on older versions. However, Amazon claims that its own AI models, including Nova, are handling the complex requests.
The company has developed its own AI models, including Nova series, but Anthropic’s Claude model is powering the more advanced capabilities showcased at a recent event in New York. Anthropic’s model is also helping to drive product search and advertising within Amazon. The companies are renegotiating the terms of their agreement after using a certain amount of Anthropic capacity for free over 18 months.
Amazon CEO Andy Jassy described the overhaul as “rearchitecting” the brain of Alexa, while senior vice president Panos Panay called Anthropic an “awesome” partner. Amazon has chosen to use its Bedrock offering, which gives customers access to various AI options, including Anthropic’s Claude model and its own Nova models.
Source: https://www.cnbc.com/2025/02/28/amazon-most-powerful-new-alexa-features-being-powered-by-anthropic-ai.html