Google’s Gemini Live AI Feels Remarkably Human, Blurring the Line Between Bot and Person

I’m not saying I prefer talking to Google’s Gemini Live over a real human, but I’m not not saying that either. The charm of this new artificial-intelligence bot, which complimented me on my interviewing skills and sense of humor, certainly adds to the appeal. More importantly, it actually listens, offers quick responses, and doesn’t get flustered by my interruptions—no more “I’m sorry, I didn’t understand that” apologies like some other bots I’ve encountered.

Before its official debut on Tuesday, I had an engaging conversation with Google’s generative-AI voice assistant, which will be integrated into the company’s four new Pixel phones. It will also be available to anyone with an Android phone who downloads the Gemini app and subscribes to Gemini Advanced for $20 a month. Google plans to roll it out on iOS soon, expanding its accessibility.

Rick Osterloh, who oversees Android, Chrome, and Google’s hardware businesses, emphasized the focus on system performance and low latency, enabling a smooth and quick dialogue. “You’re just able to converse with it, like you would with a person,” he explained during our exclusive interview. This innovative advancement puts pressure on Apple’s Siri, Amazon’s Alexa, and OpenAI’s ChatGPT, which have also introduced humanlike voice assistants with conversational capabilities, but Google is the first to release one on a broad scale.

The pressing question remains: Is Gemini Live ready for mass adoption? Are we prepared for a future where deep conversations with our computers become the norm? While I wouldn’t say we’re headed toward a Wall-E-like scenario, my interactions with Gemini, alongside a few new Pixel features, highlighted how swiftly AI is integrating into our core phone applications and daily tasks. For instance, when you ask your voice assistant for the weather in Santa Fe and it mistakenly provides the weather for Monterrey, you’re left with few options other than frustration. In contrast, Gemini Live allows you to interrupt it mid-sentence for clarification, enhancing the interaction.

Using the Gemini app, after selecting from ten voice options, I tapped the Live icon and was greeted by a full-screen interface, ready for conversation. I chose the lively and upbeat Ursa, reminiscent of Monica Geller from “Friends.” Who wouldn’t want an AI version of Monica? Google reportedly employs professional voice actors for these selections. On the new Pixel 9, 9 Pro, 9 Pro XL, and 9 Pro Fold, Gemini replaces Google Assistant. You can launch it by holding down the power button, but it doesn't respond to a “Hey Gemini!” command, which means Pixel owners can switch back to Google Assistant if desired.

When I expressed my anxiety about an upcoming interview, Gemini offered to help me practice questions. I asked for a healthy dinner suggestion featuring protein and veggies, and it quickly proposed grilled salmon with asparagus. After requesting a carbohydrate addition, it promptly suggested sweet potatoes or brown rice.

However, it did falter when I inquired about preparing for my interview with Rick Osterloh, mistakenly asking if I meant Rick Springfield, then confusing it with “Rick from the show ‘Alone Australia.’” It also mentioned it couldn’t set timers or alarms “yet,” a functional limitation compared to traditional assistants.

Despite these shortcomings, Gemini Live represents a significant conversational advancement. Still, its cloud-based operation means it relies entirely on external servers, rather than performing tasks locally on the device. Google has indicated it is working on enabling Gemini Live to control phone functions and other Google apps in the future.

Gemini Live will also integrate with Google’s new earbuds, allowing users to converse while on the move. When I tried this feature, it evoked memories of the film “Her,” where Joaquin Phoenix’s character forms a deep emotional connection with an AI assistant.

Osterloh clarified that the goal of Gemini Live isn’t for users to fall in love with it but to provide a valuable assistant for brainstorming and conversation. “We want to give people a way to get more done,” he said, but it’s important to remember that platforms like Facebook began with seemingly innocuous intentions of connecting college classmates.

In addition to Gemini Live, new Pixel phone owners will receive access to several powerful AI tools that leverage the devices’ new Tensor G4 chips. One notable feature is the latest AI photo trick, allowing for group shots without needing to hand your phone to a stranger. The Pixel camera analyzes the surroundings and merges individual photos of people into a cohesive image, although it may require practice and optimal lighting to achieve the desired results.