Wes Davis, reporting for The Verge:

Amazon is finally launching the long-awaited generative AI version of Alexa — Alexa Plus — that, if all goes well, will take away much of the friction that comes with talking to a speaker to control your smart home or getting info on the fly.

Some of the new abilities coming to Alexa Plus include the ability to do things for you — you’ll be able to ask it to order groceries for you or send event invites to your friends. Amazon says it will also be able to memorize personal details like your diet and movie preferences.

Alexa Plus is $19.99 per month on its own or free for Amazon Prime members — a better deal, considering Prime costs just $14.99 per month or $139 per year. That comes with access to the Alexa website, where the company said you can do “longform work.” Amazon also said it created a new Alexa app to go with the new assistant. Alexa Plus will work on “almost every” Alexa device released so far, starting with the Echo Show 8, 10, 15, and 21. The new system will be free for anyone in early access, which will start rolling out next month.

The actual Alexa+ announcement isn’t very interesting and was teased September 2023 as “Alexa 2.0,” a large language model-powered version of the normal Alexa chatbot. It works just like Gemini now1, drawing from a pool of personalized information and webhooks to complete more complex tasks than the Siri-like Alexa introduced in 2014. Expect it to function in practice like ChatGPT’s voice mode or Gemini on Pixel phones (formerly “Assistant with Bard”), using the breadth of the world’s knowledge thanks to web integration and various application programming interfaces thanks to Amazon’s connections and server business, Amazon Web Services.

One peculiar note is the pricing structure: $20 a month, like any other artificial intelligence chatbot on its own, but free for Amazon Prime subscribers, which, if we’re being honest, is most Echo owners anyway. Even I, an active Amazon hater, subscribe to Prime. Amazon Prime, for context, costs $16 a month and offers free Amazon delivery, among many, many other perks. This has led a lot of people to ponder how Amazon is making a profit on Alexa+ if it already made “minimal” margins on Amazon Prime by itself before running LLM servers all day for the new chatbot. The answer lies in Prime’s genius business model: it only subsidizes the free shipping aspect. Free two-day shipping to virtually any urban address, at Amazon’s scale, is dirt cheap, and the rest of the $16 is profit because the rest of the services (Prime Video, Music, etc.) are just fancy names for content licensing deals Amazon could eke out with just a few billion dollars a year.

Alexa+ is powered by Amazon’s own LLM — Amazon Nova, according to Davis — and some off-the-shelf models from Anthropic, and both are obviously run on AWS, which Amazon owns. It doesn’t have to build any new servers or pay any API costs. The infrastructure is there and has been for years — at least ever since AWS started selling AI server space for third parties after the ChatGPT boom. The $20 monthly entry fee for non-Prime members is a marketing gimmick to get people to purchase Prime, which Amazon makes a disgusting amount of profit on. It looks like a bargain on the consumer’s side, but it’s just a ploy to sell more Prime subscriptions. And if any fool actually pays the $20 for Alexa+, that’s just free money for Amazon. It’s not generous on Amazon’s part because it (a) sells more Prime subscriptions and (b) satiates investors’ appetite for AI juice in every product. Clever business move, Jeff Bezos.

This brings me to the world’s most advanced AI technology company, Apple of Cupertino, California. The new Siri — previewed nearly a year ago — better be able to develop a cure for cancer and end childhood poverty at this rate because of how long it’s been in the oven. Seriously, what the hell is Apple doing? Amazon — the company that forces its employees to urinate in water bottles and still was sweating bullets about missing its timeline — has a solution for the LLM wars while the Apple Intelligence-powered Siri has no official estimated arrival time. And forget about the truly LLM-powered one, which is aimed for a release next spring. Apple just isn’t behind; it’s living in the medieval times.

Deep inside, I know the new Siri isn’t going to be worth the wait. It’s entirely reliant on app developers to integrate App Intents into their apps, something the big names like Google and Uber aren’t bound to do anytime soon. Without App Intents — Apple’s version of Google’s Gemini apps, Amazon’s webhooks, and OpenAI’s Operator — the only Siri will really only work with Apple’s own apps like Mail and Messages. Sure, that might be helpful, but when shouting into the air at an Echo probably yields a better answer than the iPhone in someone’s hand, there’s really no point. Keep in mind: These products live together. There are hundreds of millions of iPhone users with an Alexa-powered smart speaker in their home — does Apple really think people are desperate for LLM voice assistants? If Siri fails, ChatGPT is two button presses away. Alexa+ is putting on a show, Gemini is hanging out, and Claude really is developing the cure to cancer or something like that. Nobody is waiting on Apple.

What comes first: the first public beta of iOS 18.5 containing the “new” Siri or Alexa+’s full rollout to all Echo Show devices? My money’s on the latter.


  1. According to Aaron Perris at MacRumors, Apple is poised to finally add Gemini as a chatbot alongside ChatGPT in Siri. I don’t know how this will work after Google lost an antitrust trial for doing exactly this for two decades, but I hope it includes the “apps” functionality of Gemini, which allows it to plug into Google Maps, YouTube, and Search. (Further reading from Federico Viticci at MacStories.↩︎