AI and the Rest of Us
Hello, I hope that your Fall is going great.
4 years ago I ran for President in large part on the impending arrival of AI and what it would mean. This week I spent time with Tristan Harris and Aza Raskin, the co-founders of the Center for Humane Technology. You might know them from the Netflix documentary “The Social Dilemma.” Said Tristan this week, “As a society we had a hands-off approach to social media, and there have been any number of negative effects. Hopefully we’ve learned our lesson with AI, which will be an even more dramatic and transformative technology.” They believe that government should be taking a greater role in mitigating the potential pitfalls of AI.
Also in that camp is Gary Marcus, whom I interview on the podcast this week. Gary is a professor and experienced technologist who started an AI company that was acquired by Uber. “I think that AI could improve our way of life in many ways, making us healthier and unlocking major scientific breakthroughs. But I also think there are real risks that are with us here and now.” Gary breaks down these risks in his new book, “Taming Silicon Valley: How We Can Ensure AI Works for Us.”
“AI could lead to a collapse in trust that could easily jeapordize democracy,” Gary says. “Imagine if the cost of producing bullshit is zero. And then apply that to all of the bad actors out in the world. That’s what AI is going to enable.” This is both personal and societal, as we can all be subject to audio calls and messages that sound like loved ones having an emergency but are really AI looking to scam us out of our money. Meanwhile, one can imagine countries being misled into hostilities based on bad info propagated by foreign actors.
Gary is concerned about bias baked into the algorithms. “AI is increasingly being used for hiring decisions. Imagine if certain classes or types of people are locked out of opportunities algorithmically and there’s nothing they can do. Meanwhile people using the algorithms for hiring would barely understand how they work.”
Of course, AI could change the very nature of work. “Are there jobs that will be made obsolete by AI? Yes there are,” Gary asserts. “We need a different way for the public to benefit from AI than our current tax system and a new way for people to share in that value.” Gary favors piloting income and cash transfer programs to help ease what will be an historic transition.
I’m personally very confident that AI is going to eat through many American jobs; just the other day I spoke to someone who said they used to have 15 designers suggesting various ideas for graphics and clothing. They recently fired them all because now they use AI for the same task. A lot of creative work is being changed to utilize AI, which often means fewer workers. What 4 years ago was hypothetical is now immediate.
One thing I like about Gary is that he’s no bystander; he’s already testified to the Senate about AI and is agitating for better policy. “That’s why I wrote this book, out of frustration and a deep desire to help us get this right.” He proposes a number of policies, including data rights, layered AI oversight and meaningful tax reform that would help.
It probably doesn’t surprise you that I agree with Gary’s call to action. AI is here and changing the nature of our reality and how we interact with the world. Left to its own devices there will be some major downsides. If we want more of the good and less of the bad, it will be up to us and our leaders. Time is of the essence. AI gets smarter and more powerful every day, while we are lucky to stay about the same.
To see Gary’s book click here. For my interview with Gary click here. To help advance our political system check out Forward – I will be in Denver this week at the Independent National Convention making the case for reform. A modern political system would have a much better shot at addressing AI and other 21st century challenges.
