Google tops the Index with Gemini Live and Pixel's AI features

Welcome to ZDNET’s Innovation Index, which identifies the most innovative developments in tech from the past week and ranks the top four, based on votes from our panel of editors and experts. Our mission is to help you identify the trends that will have the biggest impact on the future.

Google’s annual Made by Google product event dominates this week’s Index, but its success also caused some disagreement within ZDNET’s panel of voters. 

Coming in at #1 is Gemini Live, the tech giant’s much-anticipated elite version of its voice assistant. Pitched as a companion that gives users richer, more meandering conversations, Gemini Live can go back and forth on a topic, be interrupted, and provide advice. Unusually, the company compares the experience to being like a “regular phone call“. While the three-month turnaround on the release is impressive, its apparent multimodal capabilities remain unreleased. Google is, of course, restricting Live to Gemini Advanced subscribers with Android phones and Pixel Pro 9 owners. Otherwise, the feature is behind a $20-per-month paywall, making it only slightly more available than its competitor, OpenAI’s Voice Mode, for the moment. 

While most of our team was impressed with Google Live’s capabilities, ZDNET Editor Sabrina Ortiz disagreed that the assistant should take first place in the Index. “There is no denying that Gemini Live’s conversational capabilities are impressive, but with OpenAI’s Voice Mode also on the market, the technology doesn’t seem groundbreaking to me — yet,” she said. “Once the camera is incorporated and Gemini Live can understand your surroundings, then it will have a truly competitive edge.”

Google also comes second with its AI features for Pixel 9 series phones. With AI-powered call transcripts, intelligent interfacing with screenshots, photo-editing features, and Gemini as the default assistant, the new Pixels offer many of the capabilities Apple promised at WWDC, but without the wait. It’s still to be seen where the Pixel phones land in the on-device AI race that’s heating up between Samsung, Google, and Apple, but our experts were impressed with the features.

In third place is Nvidia, which has partnered with California to train 100,000 state residents on AI. Part of Governor Gavin Newsom’s technology-forward agenda, the program will give community colleges access to Nvidia resources to prepare students for in-demand AI jobs — and hopefully create a symbiosis that benefits California in the process. However, the bigger picture is that the initiative is part of a strategic move by Nvidia as government scrutiny of AI, and the companies working in the space, increases. Though positioned somewhat differently as a chip manufacturer, Nvidia is still making a concerted effort to show how it’s reinvesting in US infrastructure, especially at the education level, which should play well for those worried AI will imperil their jobs.

Closing out the Index is a brand-new database from MIT called the AI Risk Repository, which collects and classifies over 700 dangers posed by the technology. Geared towards everyone from policymakers to business leaders, the database aims to clarify fragmented, murky literature about the risks of implementing AI, which are numerous. The repository is the first known resource of its kind, and it’s free and downloadable. If used by policymakers, the database could play a significant role in shaping the future of AI regulation and responsibility.