Google Assistant already uses several different artificial intelligence technologies to power the convenient-yet-frustrating experience we all love to hate. Its natural-sounding voicecomes courtesy of DeepMind’s WaveNet, and its ability to recognize your commands and the context behind them stems from years of training machine learning models. But it fails to live up to our sci-fi dreams of what an AI should be, and it pales in comparison next to generative AIs like ChatGPT. That will soon change.
At its Made By Google hardware event in New York City on October 4, Google took the wraps offAssistant with Bard, blendingGoogle Bard’sgenerative AI capabilities with Google Assistant’s conversationality and tight integration with mobile devices.
Assistant with Bard amplifies the capabilities of Google Assistant by integrating it with the cutting-edge power of generative AI, a technology that can generate new content, thus proactively assisting users in novel ways. Sissie Hsiao, the Vice President and General Manager of Google Assistant and Bard, elaborated on its prowess during the keynote, emphasizing its multichannel communication avenues and real-time device assistance.
Assistant with Bard combines personalized help with reasoning and generative capabilities. So it can hear, it can speak, it can see, and it can even take actions that help you out right on the device you always have with you.
Notably, as we pointed out in a recent update,Google Bard’s Extensionsenhance the user experience even by tapping into primary apps such as Google Maps, YouTube, Hotels, and Flights. This integration is mirrored by Google Assistant, and combining the two ensures users can amalgamate data from a diverse array of sources, streamlining their creative process or simplifying basic task execution.
The Assistant’s conversational overlay feature provides a unique user interaction when combined with Bard’s generative AI. After capturing a photograph, for instance, one can initiate the Assistant with Bard overlay on it to learn more about the photo’s subject. Whether you’re trying to draft a grocery list for an upcoming weekend retreat, unearth an old email, or simply send a text, Assistant with Bard is engineered to handle these tasks with a precision and personal touch reminiscent of a human aide.
Furthermore, the assistant exhibits an unprecedented level of contextual understanding. For instance, on Android devices, if you snap a photo of a memorable moment, you may effortlessly invoke the Assistant with Bard overlay, asking it to craft a poignant social media caption. Using the image as a reference, it gauges the context and generates a fitting caption.
Bard functions akin to prominent models like ChatGPT, producing textual responses echoing human-like conversations. As a large language model, Bard ingests and processes substantial volumes of text, thus enabling natural language interactions. Moreover, its integration with Google Search ensures real-time updates, reflecting contemporary events beyond its foundational training. All of this functionality was a natural fit with Google Assistant.
Assistant with Bard is not available yet, but the wait shouldn’t be long. At the keynote, Hsiao noted “We’re rolling out Assistant with Bard to select testers shortly, and will be expanding availability as an opt-in experience in the next few months.” For comparison, Google unveiled Bardon February 6this year with a similar timetable beforeopening beta signups on March 21and eventually making it availableto anyone interested by May 10.
