Alexa+, the new generation of Amazon’s voice assistant, represents the most ambitious evolution in the eleven-year history of this product. And it comes at a good time to stop the slow decline of one of the sector’s pioneers, or at least try to.
Officially presented on February 26, 2025 and launched in early access in March, it already has more than one million active users in the United States during its first weeks.
This revamp is not a simple patch or update: Amazon has completely rebuilt the Alexa architecture, integrating generative intelligence into every phase of development, from writing code to validating responses.
Origin and initial deployment
At the presentation event in New York, Panos Panay, head of Devices and Services at Amazon, described Alexa+ as a “total re-architecture” of the assistant.
Availability was initially limited to Echo Show 8, 10, 15 and 21 users in the US, with free access during the early phase and, after the official launch in summer, at no cost for Prime members and a subscription of $19.99 per month for everyone else.
This pricing model takes advantage of the base of more than 600 million Alexa devices sold to date, seeking to make innovation profitable without fracturing the user community.
Architecture and development with artificial intelligence
VP Daniel Rausch led a development process driven almost entirely by generative AI.
Language tools like LLM acted as assistants in writing lines of code, in generating test cases, and as “judges” to rate and optimize the quality of the assistant’s responses.
Through this approach, the Amazon team accelerated iteration cycles and discovered improvements that would have taken months in traditional development. According to internal sources, the technology “Nova Act” (the underlying AI engine) was designed to raise the accuracy of complex tasks to over 90% in the medium term.
Main Alexa+ capabilities
Alexa+ adds functionalities that transform the user experience:
- Fluid and contextual conversation: interprets accents, nuances and maintains the thread of extensive dialogues.
- Personalized memory: remembers tastes, routines and specific data (recipes, events, musical preferences).
- Autonomous agents: execute complex actions such as restaurant reservations, ticket purchases or coordination of household services without additional intervention.
These advancements elevate Alexa+ from a simple command interpreter to a proactive assistant, capable of anticipating user needs.
Integration with services and devices
Alexa+ orchestrates multiple third-party APIs to expand your ecosystem: from transportation platforms (Uber Eats, Lyft) to home automation systems (Philips Hue, Ring) and ticketing (Ticketmaster, OpenTable).
In addition, the company has announced a future web version of Alexa+, accessible from Alexa.com, which will facilitate interaction from any browser.
Pending challenges and functionalities
Despite its robustness, Alexa+ still does not include all the features initially announced.
Features like food ordering, interactive Fire TV control, and browser shortcuts continue to be developed, with accuracy rates on two- or more-step tasks still above 60% in internal testing.
Other delayed capabilities include the generation of personalized stories for children and visual reminders based on facial recognition, which are expected to arrive in the coming months.
User perception
The first evaluations have been mixed. A part of the community praises the naturalness of conversations and the efficiency when managing smart homes, noting that Alexa+ makes the user feel “as if they were talking to a real person.”
However, some users report incompatibilities with older devices and describe certain processes (such as chaining multiple home automation actions) as “somewhat rough” in their execution.
These criticisms highlight the challenge of maintaining backward compatibility while introducing innovations that require computing power and new architectures.
Competitive strategy
Alexa+ lands at a time of intense rivalry between voice assistants. In 2024, OpenAI presented the “Advanced Voice Mode” integrated into ChatGPT, capable of maintaining fluid dialogues and executing voice scripts; For its part, Apple is preparing a major Siri update for 2026 with renewed AI.
Google, meanwhile, offers Gemini on Nest devices and Android phones. Amazon’s bet combines the scalability of its network of Echo devices and Prime’s drive to retain users in its ecosystem.
Roadmap and future perspectives
Amazon has advanced several lines of work for the coming months:
- Global rollout: After the US, Alexa+ will arrive in Europe and Asia before the end of the year.
- Open SDK: release of development kits for third parties to integrate their services and devices with Alexa+.
- New devices with NPU: Next-generation Echo Show with dedicated AI processors to reduce response latencies.
These improvements seek to ensure that Alexa+ is not only smarter, but also faster and more accessible in any environment.
Alexa+ represents a before and after in the evolution of voice assistants, thanks to its full integration of generative intelligence, its agentive capacity and its focus on a more human and personalized interaction.
Although challenges persist in compatibility and full functionality, the project is advancing steadily: it already exceeds one million users in Early Access and is preparing its global expansion, positioning itself as one of the most powerful competitors in the conversational AI market.
As Amazon refines the technology and delivers on remaining functionality, Alexa+ is set to redefine the way we engage with technology at home and on the go.
This post is also available in: