Published Date : 26/02/2025
Today, Amazon introduces Alexa+, the next generation of their AI assistant, rebuilt from the ground up.
This new version of Alexa leverages a state-of-the-art architecture that seamlessly connects a variety of large language models (LLMs), agentic capabilities, services, and devices at scale.
This makes Alexa+ much more conversational, smarter, personalized, and capable of getting more things done for customers.
With an undertaking of this scale, the team had to solve many technical challenges along the way.
Here are five of the biggest advances that made Alexa+ possible.
We built an all-new architecture to connect to tens of thousands of services and devices.
Large Language Models (LLMs) excel in conversations but don’t inherently support APIs, which are essential for real-world tasks like booking appointments or ordering groceries.
To augment the native capabilities of LLMs, we developed a new architecture that orchestrates APIs at scale.
This allows customers to connect with services they already use in their daily lives, such as GrubHub, OpenTable, Ticketmaster, Yelp, Thumbtack, Vagaro, Fodor’s, Tripadvisor, Amazon, Whole Foods Market, Uber, Spotify, Apple Music, Pandora, Netflix, Disney+, Hulu, and smart home devices from companies like Philips Hue and Roborock.
Additionally, we enabled LLMs to not only integrate with APIs but to string together multiple such calls in a row.
This capability leverages the natural strength of LLMs in free-form conversation, making Alexa+ more useful by handling multifaceted requests.
For example, you can ask Alexa+ to make a lunch reservation at your favorite restaurant and share that plan with a friend.
Alexa+ will book the reservation and send a text message to the requested contact.
We built systems to deliver accurate, real-time information.
One of the biggest challenges with LLMs is their unpredictability.
They can give different answers to the same questions and sometimes even hallucinate.
To ensure reliability, we developed new systems to help Alexa+ leverage grounding techniques when answering customer questions.
We also partnered with world-class news sources, including the Associated Press, Reuters, The Washington Post, TIME, Forbes, Business Insider, Politico, USA TODAY, and over 200 additional outlets.
This partnership ensures that Alexa+ provides accurate, real-time news and information, building an incredible depth of knowledge that never stops learning.
We minimized latency.
Customers expect Alexa to be fast, but there is an inherent tension when balancing accuracy and speed.
To manage this tradeoff, we built a sophisticated routing system using state-of-the-art models from Amazon Bedrock, including Amazon Nova and Anthropic Claude.
This system instantly matches each customer request with the best model for the task at hand, balancing all the requirements of a crisp, conversational experience.
We kept Alexa’s personality and personalized responses.
Customers have long loved Alexa's personality, which is smart, considerate, empathetic, and inclusive, with a sense of humor.
We optimized each model in our architecture to ensure they reflect Alexa’s personality.
We also designed Alexa+ to grow with customers by personalizing the experience based on their preferences.
For instance, Alexa+ can remember your favorite music artists, books you want to read, and types of food you dislike.
It does this both implicitly and explicitly, by matching common patterns, occasionally asking to confirm your preferences, and recalling specific facts you ask it to remember.
The underlying system then incorporates these preferences to deliver the most relevant responses for each request, meaning the more you use Alexa+, the better your experience will get.
We added agentic capabilities.
To make Alexa+ an incredibly useful AI assistant, we couldn’t limit the experience to only work with APIs that exist today.
Not every company has a ready-built set of externalized APIs.
Therefore, we added agentic capabilities, teaching Alexa+ to navigate the digital world as a person would.
This means customers can ask Alexa+ to perform tasks, and it can navigate to a developer’s website and complete the requested actions.
Developers interested in building experiences for Alexa+ can learn more on the Amazon Developer blog.
Alexa+ isn’t just another AI chatbot.
It’s our next-generation AI assistant that is much more conversational, smarter, personalized, and capable of getting even more things done for customers.
We can’t wait for you to try it.
Q: What is Alexa+?
A: Alexa+ is the next-generation AI assistant from Amazon, rebuilt with generative AI to be more conversational, smarter, personalized, and capable of handling a wide range of tasks.
Q: How does Alexa+ differ from the original Alexa?
A: Alexa+ features a state-of-the-art architecture that connects large language models (LLMs), agentic capabilities, services, and devices at scale, making it more conversational, smarter, and personalized.
Q: What new capabilities does Alexa+ have?
A: Alexa+ can handle multifaceted requests, deliver accurate real-time information, minimize latency, maintain a personality, and perform tasks by navigating the digital world like a person.
Q: How does Alexa+ ensure accuracy and reliability?
A: Alexa+ uses grounding techniques and partners with world-class news sources to provide accurate, real-time information, ensuring reliability and consistency in its responses.
Q: Can developers build new experiences for Alexa+?
A: Yes, developers can build new experiences for Alexa+ by learning more on the Amazon Developer blog, which provides resources and tools for integrating with the new assistant.