When we started with digital assistants, Microsoft potentially had the inside track. Its Cortana effort was named after the Halo AI character, which gave it far more potential than the others.
Apple had a decent shot, too. It marketed Siri very well but didn’t seem to want to fund its advancement, while Amazon just shipped product after product that used Alexa.
Google was no slouch either, pivoting hard to its Gemini product (which I use a lot) for Pixel smartphones.
What I find fascinating is that Panos Panay, the guy who pretty much made the Microsoft Surface line a thing and drove my favorite phone, the Surface Duo, is now at Amazon and apparently doing there what should have happened at Microsoft with Cortana.
Let’s discuss how we got here, what to expect, and why this move from Amazon will be a game-changer. We’ll close with my Product of the Week: AMD’s new RX 9070 and RX 9070 XT graphics cards.
Microsoft’s Cortana Failure
Microsoft was one of the first to introduce a believable AI assistant. What made it believable was that it was named after a character in the Halo video game. To make it work, Microsoft needed to advance this technology and make it look like it did in the game.
You might have thought this was a stretch, but an engineer working on his own created exactly what this would have looked like, and it worked. Imagine how this would have set the stage and provided a huge competitive advantage for the company once generative AI was driving into the market years later. It would have been the centerpiece of any tech-forward home, particularly for those of us who got hooked on Halo back in the day.
Then, instead of becoming the digital assistant that would rule them all, Microsoft did what it so often does these days: it underfunded and under-executed the effort with virtually no help from marketing and eventually pulled the plug on it years before generative AI became a thing.
Apple’s Siri Misstep
Apple also had a shot at this race with Siri, which was, compared to Microsoft, far better marketed and executed even though, like Cortana, it really wasn’t AI so much as a text-to-speech, speech-to-text interface for the web. Apple had more people using Siri than any of the alternatives, but the promise of Siri that the ads tended to convey was well short of Siri’s reality.
Even today, when I’m on a panel discussing AI, someone almost always points to Siri and argues that it isn’t AI because it sets such a bad example.
Apple had the opportunity as part of its now failed partnership with IBM (who saw that coming?) to have Siri be the front-end for IBM’s watsonx, which would have made Siri into the first digital assistant that was a true AI. But that partnership was virtually dead on arrival, largely because Apple just doesn’t work well with partners. While IBM has a far better partnership skill set, even IBM couldn’t save the thing.
Siri remains in use — though that may change now that Apple can spell AI — but it’s still the disappointment it has always been.
Alexa Surprise
Amazon has lost billions on Alexa. However, arguing that Alexa is a failure is unfounded mainly because many users order products from Amazon using Alexa. Unlike the others, Amazon saw the potential to create a front end for the retail business and use the digital assistant to remind and help people to order stuff and become a trusted part of their day.
I know we often start our day asking Alexa about the weather and end the day asking Alexa about actors in shows (for those not using Amazon X-Ray, which is a truly handy feature available with Amazon streaming content) or getting color on the limited information we’re watching on TV.
Even though, just like Cortana was, and Siri is, Alexa is more of a web front end than true AI and has proven far more useful. The only thing we use almost as much is Google’s digital assistant, which, for some reason, doesn’t have a cute name.
What will change in the coming weeks is that Alexa moves to generative AI. Granted, that move was promised in 2023, but this turned out to be more difficult than they thought. Still, it is better to do something well than quickly — and while this feature will only be for Prime members, so it is not truly free, the result should be a huge improvement in this technology class.
The generative AI assistant will be called Alexa+. It will work on most Alexa devices, a smartphone app, and the Amazon website. So, although Amazon wasn’t first, didn’t have a game to connect to, and was more of a retailer than a tech company, it out-executed Apple, Google, and Microsoft.
Google is arguably a distant second with its planned implementation in Pixel phones, which may be a problem for me because I carry a Pixel phone, and I might prefer Alexa+.
Wrapping Up: Alexa+
If you aren’t working with AI, you should start because you’ll need that skill set to do many things in the second half of this decade. AI can handle complex questions with detailed answers, generate content, offer nuanced advice, and even assist with productivity.
Alexa+ will let up jump to scenes in movies you own or rent on Amazon’s video service, ask for unique playlists on Amazon Music, and get far better answers than just a reading of a web page that Alexa would have previously defaulted to.
You’ll be able to use natural language rather than the Alexa commands you need to learn, and you’ll be able to give it complex, multi-step directions based on historical knowledge. For example, you can ask it to book the same place for your anniversary that you booked last year, purchase tickets at specified prices, or even recommend and contact repair services for something broken.
With the Rink security cameras — and this is huge — you could ask Alexa+ to show you any images of people who have been frequenting your property, how often your kids are actually walking the dog, or to figure out where your cat is (assuming a camera has or can see it).
There will likely be some quality problems because AI developers aren’t focusing enough on quality yet, which I think is a huge mistake. So you’ll want to be careful about taking everything Alexa+ says as fact. There will likely be some issues with third-party applications until they become fully integrated, but this is huge, and I think it will change how we look at digital assistants going forward.
Panos Panay wasn’t able to execute at Microsoft, though I know he tried — so it’s great to see Amazon give him what he needed to perform there. Maybe a future Amazon device will be able to do what that Cortana prototype did and give Alexa+ a face. We’ll see.