By Tim Paradis
Famed Silicon Valley investor Vinod Khosla thinks artificial intelligence will be able to do 80% of the work of 80% of jobs.
That will mean humans have a lot more free time because the value of our labor will fall. One way to cushion the blow: universal basic income.
Khosla, who cofounded Sun Microsystems and who has invested in OpenAI, wrote in a lengthy post on the website of his eponymous venture capital firm that AI would reduce costs and make expertise nearly free.
That means everyone from doctors to those in sales to people who work on farms and assembly lines could see AI take over much of their work — and mostly do it better, he wrote.
“As AI reduces the need for human labor, UBI could become crucial, with governments playing a key role in regulating AI’s impact and ensuring equitable wealth distribution,” Khosla said.
Advocates of universal basic income, including Silicon Valley luminaries like Elon Musk and OpenAI’s Sam Altman, have said it could help blunt disparities that threaten to widen as AI absorbs aspects of jobs — or replaces some roles altogether.
AI isn’t like other technological advances
Khosla, who described himself as an “unapologetic capitalist and technology optimist,” contends that widespread payments to supplement income will likely be necessary, at least for a time, because AI isn’t like many recent technological advances.
The microprocessor, the internet, and the mobile phone emerged as tools humans could use, he said, but “AI, by contrast, amplifies and multiplies the human brain much as the advent of steam engines and motors amplified muscle power.”
He expects AI to drive down costs well beyond what the microprocessor could. In addition to making all expertise broadly available and nearly free, Khosla expects AI will usher in everything from bipedal robots to cheaper materials — including metals and drugs by supercharging science and the discovery of materials.
He understands concerns over the idea of handing out cash to people who are pushed out of jobs by AI. Khosla wrote that it might seem “impractical” because of economic constraints. And, he added, ignoring those problems had led to “disasters” in places like Argentina and Venezuela.
Yet Khosla expects the breadth of gains from AI will gradually reduce the economic constraints that animate some criticisms of UBI.
One of those is that UBI programs could remove an incentive to work. Since 2019, more than 100 municipalities in the US have tried a version of handing out money. So-called guaranteed basic income programs are focused on lower-income households as a means of fighting poverty.
A recent study found that rolling out a worldwide UBI would increase global GDP by 130%. Last week, Pope Francis reiterated his support for UBI because, he said, it could help lift some people who haven’t seen the benefits of globalization.
Discussion is needed
Khosla also said it’s important to be careful when making recommendations for any particular policy fix or premature moves at a national scale that would be drastic or irreversible.
“Debate and discussion are definitely called for,” he said.
Yet interventions are needed to help those hurt by broadening income disparities, Khosla said. That will mean watching these changes closely and making “small policy changes” in this decade.
He noted that even if AI continues to make rapid gains in its abilities, the impact and adoption might be slower, “much like the flat part of an exponential curve.”
A billion robots
Khosla predicted that robots will produce sufficient value to “support the people they replace.” He said that in 25 years, there could be one billion two-legged robots handling various tasks.
That could eliminate the need for people to handle sometimes-grueling tasks like working on an assembly line or a farm, and such a market could become larger than the auto industry, Khosla said.
“Few are preparing for how this will radically change GDP, productivity, and human happiness and free people of the servitude of these assignments we call jobs,” he wrote.
