AI is a hot topic on Wall Street—but it’s also one of the most divisive. Experts in the field, including Big Tech leaders, start-up founders and computer scientists, can’t seem to agree on whether the technology—which sparked an investing frenzy after OpenAI’s ChatGPT became a mainstream phenomenon—will save humanity or destroy it.
Only one thing is likely, says Peter Orszag, one of President Barack Obama’s former top advisors and now a top Wall Street CEO: they’re all probably wrong. Speaking at at Fortune’s Global Forum in Abu Dhabi on Wednesday, the Lazard CEO hazarded a few predictions himself, but only after he said: “With any big new technology, comments that are made at this kind of stage with regard to what then subsequently happens are invariably wrong.”
Orszag, who served as director of the Office of Management and Budget and director of the Congressional Budget Office in the Obama administration—said a “huge dose of humility [is needed] here in terms of how this will play out, because if you look back at the introduction of a whole variety of technologies, in terms of how they were predicted to affect anything … that track record is terrible.”
Of course, Orszag, who has officially been Lazard CEO for about a month, is wading into the debate between the “accelerationists” and the “doomers,” which captivated Silicon Valley and the business press during Sam Altman’s five-day expulsion from OpenAI.
Utopia or armageddon?
Some of the more extreme views on the dangers of AI—from the likes of Elon Musk, Apple co-founder Steve Wozniak and Geoffrey Hinton, one of the so-called “Godfathers of AI”—have included anxieties over superintelligent computers “going terminator,” attempting to overrule humans, or being used as killing machines on battlefields.
At the other end of the prediction scale are Microsoft CEO Satya Nadella and billionaire venture capitalist Marc Andreessen, who have predicted that the technology has the potential to create a “utopia” and “save the world.”
Orszag, once dubbed a “scholar-banker” by the Financial Times, was named the successor to the outgoing Lazard CEO earlier this year as the firm was contending with a dealmaking downturn. Weeks before taking the reins, he publicly set a goal for the company to double its income by the end of the decade.
Asked how he believed AI would reshape productivity in the U.S. and beyond, Orszag said the tech was likely to take a much bigger toll on hard skills than soft skills.
“So I think, in medicine [for example], that AI will more quickly replace the radiologist than the primary care doctor, where human interaction is more important,” he speculated.
One question that remained largely unanswered, according to Orszag, is how the AI revolution will affect the sharing of information.
“A core crucial thing that no one has a good answer to is what does this revolution do to what I’ll call ‘truth,’” he said. “We have not yet seen the full impact of all of these tools being unleashed in terms of their output, being then put back into the public domain. As that happens, the feedback loop from the well-known hallucination problem will become more severe.”
He predicted that it would become more difficult for media outlets to establish themselves as sources of truth (including Fortune itself)—but it would simultaneously become more important to have trusted news sources in existence.
“It’s going to be harder to do that, because you can’t just click through to the underlying website if that’s then getting infected by the output of the tools themselves,” he said.
Breakthroughs ‘you didn’t imagine’
Jenny Johnson, president and CEO of global asset management firm Franklin Templeton, also shared her thoughts on the AI revolution during Wednesday’s onstage discussion.
She reflected on the disruption brought by the dawn of the internet, noting that we “all have things that we never knew we needed that are now at our fingertips.”
“Whether you measure that in productivity or just being able to enjoy more things in life, I think that the internet did bring that,” Johnson—who was named one of this year’s most powerful women in finance by American Banker—said.
She speculated that AI, like many technological breakthroughs throughout history, would make industries including wealth management more efficient, but also give people higher expectations of services and businesses.
“That will eat up some of those efficiencies, because for the same price point we’re providing way more services [that] we probably receive less for now,” she said. “Do you measure that in productivity? I don’t know. I measure what our clients achieve or get for the dollars they spend with us, and I think with AI there are going to be a lot of [increased] efficiencies in our operational area.”
Ultimately, Johnson, who helped build Franklin Templeton into a powerhouse that manages assets worth around $1.5 trillion, said she was excited about the next technological phase that would “break through on the things you didn’t imagine”—and that like Orszag, she was skeptical about some of the wilder predictions being made about AI.
“Remember, in my business—where we use financial advisors to distribute our funds—[one magazine] had on the cover: ‘The death of the financial advisor,’ ‘The death of the broker,’” she recalled. “Well, no, it just made the broker way more efficient, and now the client expects way more services from that broker. I think that, ultimately, is the next phase.”