Amazon is discontinuing its standalone Rufus shopping chatbot and repositioning its Alexa assistant as the central pillar of its artificial intelligence shopping strategy.
The company has formally launched the Alexa Shopping Assistant, an e-commerce AI tool designed to answer customer queries and execute shopping tasks on their behalf. Amazon states this new tool integrates capabilities from both the Rufus project and Alexa+, leveraging a user's purchase history and other data with the goal of creating the world's best and most personalized AI shopping assistant.
As part of this strategic pivot, Amazon is integrating Alexa directly into its platform's search results pages. When users browse products, a chat window will automatically appear, displaying product information and offering personalized recommendations.
Amazon initially launched Rufus over two years ago, during the surge of generative AI interest across the tech sector, positioning it as a core feature on its website and mobile app to compete in the AI arena. While continuously developed, the Rufus chatbot remained in a beta testing phase.
Although the standalone Rufus chatbot is being shut down, Amazon confirms it will retain the product recommendation algorithms and user shopping history data from the project to power the new Alexa Shopping Assistant. Users can access the assistant via the stylized 'A' icon on Amazon's website and app, or through Amazon Echo smart displays.
The rise of AI shopping agents is reshaping the e-commerce landscape, prompting Amazon's strategic realignment. Over the past year, companies like OpenAI, Alphabet, and Perplexity have launched smart search tools and shopping agents, challenging traditional online shopping models. However, many such products have faced adoption hurdles, and it remains unclear if consumers are ready to fully delegate purchasing decisions to AI.
Daniel Rausch, an Amazon Alexa executive, stated that this new shopping assistant holds advantages over other AI shopping tools in the market. It can access exclusive core data such as authentic user reviews and a vast product catalog, and can provide precise information on product inventory status and estimated delivery times.
"In using it ourselves, we understood why other AI products struggle in the shopping domain," Rausch said in an interview. "Shopping isn't simply about scraping web information and inserting it into a conversational template."
Earlier this year, OpenAI significantly adjusted its own AI shopping strategy, shutting down a one-click checkout feature within ChatGPT and instead partnering with major retailers to build dedicated shopping applications within its chatbot. OpenAI stated that dedicated apps would provide a smoother purchasing process for users.
Rausch said he is not surprised that other companies have withdrawn various incomplete and disjointed AI shopping modules.
"These features simply aren't worth the cost," he remarked. "Shopping is never a trivial side task."
Amazon has been reluctant to partner with competing AI platforms and has refused to grant external shopping agents access to its platform. Amazon CEO Andy Jassy has stated that while the company is in talks with third-party AI agent services about potential collaboration, it currently blocks a significant number of external bots from accessing its site.
Concurrently, Amazon has launched a "Buy for Me" feature, which uses AI to purchase items for users, even from other retailers' websites. This rollout has drawn criticism from some retailers who claim they never voluntarily opted into the service.
Embedding the Alexa Shopping Assistant into search results pages also secures a valuable promotional real estate for Amazon.
This move could impact the platform's millions of third-party sellers, who traditionally invest significant funds in advertising to compete for top spots in conventional search rankings. The product promotion ads referenced by Amazon are a major source of its advertising revenue.
Rausch stated that the Alexa Shopping Assistant will incorporate relevant advertisements in appropriate contexts, but only where they enhance the shopping experience, and will not deliberately narrow search results.
"In some shopping scenarios, the assistant can actually show users more relevant products, depending on what stage of the decision-making process they are in," he said.
Comments