Google's Fei-Fei Li: Vision is AI's "killer app"

Computer vision is "one of the most important elements of machine intelligence and the transformation of enterprise and companies," Li said at the Google I/O conference.
Written by Stephanie Condon, Senior Writer

Artificial intelligence is "the driving force of the fourth industrial revolution," Google's Fei-Fei Li argued during a panel discussion on the final day of Google's I/O developer conference. AI, she added, "has the potential to tranform the way humans live, work, communicate."

That said, technologists have only just begun realizing the impact of AI. As it matures, vision is going to be the "killer app of AI," said Li, the chief scientist of AI and machine learning for Google's cloud unit.

Reiterating points she recently made to TechCrunch, Li noted that 540 million years ago, vision was likely the driving force behind the Cambrian Explosion -- a relatively short period of time over which a great diversity of life forms appeared. "When eyes were first developed in animals, suddenly animal life becomes proactive... evolution has changed," she explained. Hundreds of millions of years later, humans are the most intelligent visual animals, with half of the human brain dedicated to visual processing.

Meanwhile, the sheer scale of visual data in our current world speaks to its importance: It's estimated that more than 80 percent of all of cyberspace is in pixel form, Li noted. Google Cloud Platform customers, she noted, are asking about capabilities like video processing and video analytics.

Computer vision, she stressed is "one of the most important elements of machine intelligence and the transformation of enterprise and companies."

While the last 10 years have brought significant progress in basic perception tasks, such as object recognition and image tagging, Li called the next big phase "vision plus X." In other words, the transformation of enterprise will come with combining the capabilities of computer vision with other fields of study.

For instance, she noted that vision plays a "fundamental" role in communication, opening up opportunities to combine vision and language capabilities for functions like indexing videos. Vision also has huge potential in biological sciences and robotics, she said.

At I/O, Google showcased its advances in computer vision, which are helping to drive its AI-first strategy. The company revealed how it's bringing computer vision capabilities to various products this year, starting with Google Assistant and Google Photos.

Artificial intelligence has reached its current inflection point, Li argued, because of the convergence of three factors: the maturation of machine learning tools, Moore's Law and the improvement of hardware and the emergence of big data.

While this is a "historical moment" for AI, "it's just the beginning," she said. "The tools and technologies we've developed are really the first few drops of water in the vast ocean of what AI can do."

Daphne Koller, chief computing officer at Google-founded Calico Labs, noted that big data is not yet as ubiquitous as some may think.

"We might like to think big data is, at this point, the solution to everything," she said on the panel with Li. While it plays a major role in some sectors, "in others we are unfortunately still in the medium or even small data regime."

With that in mind, she said there's still a need for "balancing human intuition... with the data we're acquiring."

Editorial standards