I’ve been wearing smartwatches for nearly a decade now. And so far, I have cultivated a love-hate relationship. The fashion police in my friends’ circle abhors them, while the gym rats are always eager to discuss the latest biosensing breakthroughs. Doctors and researchers tell me that they are a fantastic tool for staying in shape and as an early warning system, but not a substitute for proper medical care.
But as smartwatches kept getting faster and smarter over the years, so did their distraction potential. “Yet another screen on my body, a channel for distracting notifications, and something I need to charge every day.” That’s what I often tell myself, and I’m not the only one grappling with those problems. But here’s the bright side.
Practical conveniences always trump the functional hiccups, and that’s where AI comes into the picture. AI has dramatically changed how we browse the web, or get work done on our phones. Now, it has landed on smartwatches, and it’s phenomenally useful. But here’s the odd part.
Apple Watches are usually deemed the gold standard, but it’s Google’s Wear OS powering the Pixel and Galaxy Watch that has taken the early lead with meaningful AI-powered experiences. I recently pushed the new Apple Watch Series 11 and the Galaxy Watch 8 Classic, and the difference in AI-driven features was quite eye-opening.
Why do I want AI on my wrist?
Well, the most obvious answer would be a digital assistant that can handle mundane tasks that would otherwise require you to unlock your smartphone. Checking the meeting schedule for the day ahead, finding a nearby coffee shop, looking for directions, changing the music mood, or just a quick news run for hot events.
The problem, however, is that legacy digital assistants such as Alexa, Siri, Google Assistant, and Bixby are extremely limited. They can’t interact with other apps or third-party services. They are pretty slow and lack contextual understanding. And above all, the conversations with them are extremely rigid.
That’s where Gemini, with its advanced natural language capabilities, makes an evolutionary leap. It can carry on conversations with a lot more nuanced understanding than Siri. More importantly, it can dig within the trove of your data across connected Google (and third-party) services to take action. And by action, I mean meaningful work on the move.
Gemini is a lot better at talking (and understanding)
Gemini is a large language model that can handle conversations almost like a human, but with a whole world of knowledge at its fingertips. It can understand multi-step queries and open-ended conversations, while Siri would either fail or try to take you on a web search journey.
The conversations with Gemini are free-flowing, and more importantly, a lot less error-prone. The latter is extremely important. Even when the voice-to-text transcription is not accurate and you see a clear spelling error on the screen, the context is picked up accurately.
For example, when I asked Gemini to give me the latest update on Nvidia’s Rubin GPUs, Gemini initially picked it up as “Reuben,” but when it finally started narrating the answer, it correctly focused on the upcoming Rubin architecture. This context-based correction also means the voice chats are a lot less frustrating on the Galaxy Watch 8 compared to the Apple Watch Series 11.
For people who don’t speak with a native English accent, it’s often a struggle with their unique pronunciations of the same words. Gemini fares far better than Siri at this crucial convenience. Siri repeatedly failed at picking up the names of places and delicacies in Indic languages, while Gemini’s enhanced language understanding allowed it to pick the right word.
Understanding the language (with its diverse accents), fixing spelling errors in real-time, and picking the right context even if users fumble with their sentences is a huge win for Gemini. Siri conversations can quickly get frustrating because it repeatedly fails to grasp words that are non-English.
And let’s not forget that Gemini is just better at answering your burning queries. Unlike Siri, which tries to answer by relying on web search and providing links, Gemini is just more spontaneous. It is dramatically faster at collecting information from the web and presenting it in an easy-to-grasp nugget of text, accompanied by a voice readout.
From there, you can dig deeper with follow-up questions, ask it to pull up an article for in-depth reading, save the contents of the conversation in Keep, or send it as WhatsApp text. Simply put, it’s a straight chain of commands that follows a fruitful back-and-forth conversation.
Command and execution
I’ll get straight to the point here. “Tell me about a few famous places in Dharamshala,” I asked. Ideally, you would expect an AI assistant to read out a few places or provide an answer on the screen. Alas, that’s not the case with Siri. When I asked the aforesaid question, it directly launched a Map view of a guest house in the city, but that was not even my query in the first place.
That’s simply an erroneous understanding of a user query and a wrong task execution, as well. When I pitched the same question to Gemini on the Galaxy Watch 8 Classic, it provided a summary of tourist hotspots. But more importantly, it saved the context of the conversation.
When I asked it to “tell me more about the Buddhist temple” without giving any specific context, it accurately gauged that I was talking about the one in Dharamshala, and even provided a picture. Next, I asked it to save the list to my Google Keep directory.
Gemini not only picked up an appropriate title for the new digital note, but also successfully saved it to my Google Keep dashboard. Likewise, it did a reliable job with most of the apps and services that are available within the Google Workspace bundle. Gmail, Maps, Calendar, and more.
Even with third-party services such as Spotify and WhatsApp, Gemini handled the requests really well. Siri, on the other hand, struggled with language comprehension on too many occasions, and the lack of interplay with third-party only worsened the situation.
And it’s not just the services that are unavailable for watchOS, but also those that are installed on the smartwatch itself. For example, I have Outlook installed on the Apple Watch Series 11, but Siri can neither access its content, nor pull any context from the inbox.
When I asked Gemini about “the last email conversation I had with Saba,” it accurately pulled up details from the exact email I wanted, and also provided a quick summary of the exchange. Likewise, Gemini also fared well with giving me a recap of the meeting schedule and setting up new calendar events.
Overall, Gemini — running atop Wear OS on the Samsung Galaxy Watch 8 Classic — is a far superior AI experience than Siri on the Apple Watch Series 11. It’s unclear how long it will take for Siri’s rumored AI brain transplant to land. And even if it does, I am skeptical about its ability to catch up with Gemini, especially with the deep level of integration it has to offer with other apps and services.
