Amazon's AI shopping assistant doubled its active users during the second quarter, while interactions jumped more than fivefold from a year earlier, the company reported Thursday. U.S. customers who use Alexa for Shopping spend more than 40% more per order than those who don't, CEO Andy Jassy said on the earnings call. The numbers show how agentic tools are already changing sales outcomes - not just assisting, but actively driving higher basket sizes.
Alexa for Shopping combines the Rufus recommendation engine and Alexa+ to compare products, surface price histories, and automate purchases through price alerts and Auto-Buy. The assistant acts as a persistent shopping agent, nudging customers toward buying decisions without requiring a human sales rep. For sales teams, the message is clear: AI that recommends, compares, and closes is moving from pilot to production at enormous scale.
AI agents move into business workflows
Amazon is also placing agents directly inside enterprise operations. Jassy highlighted Amazon Quick, a new agent that can search email, calendars, files, and company systems, then take actions - scheduling meetings, sending messages, updating customer records, and building dashboards. Users can create autonomous agents in plain language to handle multistep assignments in the background.
The tool grew from Amazon employees' own demands. Quick initially helped with document summaries and research. Workers then pushed the company to connect it with email, Slack, and calendars. "It's pretty remarkable not only how fast it's taken off inside Amazon, but how many external enterprises have put it into production with a very large number of people at their companies," Jassy said. For sales teams, an AI Agents & Automation tool like Quick could handle CRM updates, meeting prep, and follow-up tasks that currently eat into selling time.
Contact centers and autonomous transactions
Amazon Connect, the company's contact-center platform, is used by major airlines, banks, and healthcare companies and continues to grow quickly. Jassy said Amazon sees "a very substantial opportunity" in building agentic applications for customers and AWS. The company also added payments to Bedrock AgentCore, letting agents execute transactions autonomously - a capability that could reshape how sales are closed within automated workflows.
Frontier models and the multimodel bet
When an analyst asked whether Amazon needed its own leading AI model, Jassy's answer was direct: "There is not going to be one model to rule the world." Different models will lead at different times, he said, and companies building important AI applications will want access to multiple options. That puts the emphasis on Bedrock's selection, price, security, and governance rather than on owning the top model at any given moment.
Still, Amazon is developing its own frontier model, which Jassy said would give the company more control over costs for consumer applications and help lower costs for AWS customers. Bedrock's momentum backs the strategy: customers spent more on the service during Q2 than in all previous quarters combined. Amazon's AI business and chip business each passed annual revenue run rates of $25 billion, with both growing at triple-digit percentages.
Why this matters for sales
Amazon's agentic push shows that AI is no longer just a back-office tool. AI for Sales is moving into the buyer's journey itself - recommending products, comparing options, and triggering purchases automatically. The 40% higher per-order spend among Alexa for Shopping users isn't a minor uplift; it's a signal that well-built agents can meaningfully increase revenue per customer.
For sales professionals, the takeaway is practical. Tools like Amazon Quick can reduce administrative drag by handling scheduling, CRM updates, and research. Contact-center agents can field routine inquiries, leaving human reps to focus on complex deals. And autonomous transactions through Bedrock AgentCore hint at a future where the close itself becomes automated. The companies that embed these capabilities early will have more time for high-value conversations and strategy - the work that actually moves pipelines forward.
Your membership also unlocks:
