Google, just like every other tech company out there, is focused on becoming an artificial intelligence company.
So what’s the biggest problem facing AI innovators everywhere?
According to Jeff Dean, head of the machine-learning team Google Brain, one of the company’s key research areas within the 1,000-strong Research at Google team, it’s diversity.
Speaking as part of a Reddit Ask Me Anything, when asked if he thought an AI apocalypse was linked to a lack of humanistic thinking, and a lack of diversity, he said:
“I am personally not worried about an AI apocalypse, as I consider that a completely made-up fear. There are legitimate concerns around AI safety and policy, and our group (in collaboration with a number of other organizations) has recently published an Arxiv paper about some of these (see Concrete Problems in AI Safety ). I am concerned about the lack of diversity in the AI research community and in computer science more generally.”
Of a team of around 35, only three of the Google Brain researchers are women. There are, in fact, as many men called Jeff (OK, including Geoff) on the team.
There are perhaps 10 black or brown faces, however, using my rudimentary human intelligence, it isn’t easy to tell if any are from the LGBTIQ or disabled communities.
Dean is clearly aware of the shortcoming this creates, particularly as his team’s whole focus is to ‘make machines intelligent. Improve people’s lives’, the latter part of which is focused on creating good human outcomes.
Although lacking in much diversity, his team is interdisciplinary, with physicists, mathematicians, biologists, neuroscientists, electrical engineers, computer scientists and even has a philosophy grad.
Why is diversity important to him?
“In my experience, whenever you bring people together with different kinds of expertise, different perspectives, etc., you end up achieving things that none of you could do individually, because no one person has the entire skills and perspective necessary.”
He wants diversity not least because his team will perform better, particularly with more women in it, but because women are also more empathetic, which lends itself well to the social mission of Google Brain, and also because diverse voices add valuable experience.
The AI, it seems, is the simple part.