Big tech is panicking about AI and terrifying young workers, says Microsoft’s Brad Smith
The three-decade Microsoft veteran is pushing back against what he describes as a culture of manufactured alarmism in the tech industry. Brad Smith, who serves as the company’s vice president and president, says influential voices in the tech world are consistently wrong about the speed of disruption, consistently hypocritical about the solutions they propose, and consistently inattentive to the human cost of the panic they generate.
His remarks, made at Microsoft’s campus in Redmond, Washington, come alongside a newly published paper on what artificial intelligence will really mean for the workforce in the coming decades.
“No one knows for sure, but let’s not panic,” Smith said.
Graduation Season Sent a Warning to the Tech Industry That Shouldn’t Be Ignored
Smith argues that the clearest recent evidence of a fractured relationship between the tech industry and the public came not from a boardroom or policy debate but from college stages across the country this spring. He was booed when artificial intelligence was mentioned at graduation ceremonies. Smith believes the industry should just get by with it.
The backlash should be “a powerful wake-up call for the tech industry,” he writes in his article. He goes further: “We hope our industry’s leaders will listen to this backlash and try to learn from it. Over the past half-century, the youngest generations of people and workers have led the way in adopting new digital technologies. When people using a new technology complain about it, we better take notice.”
The signal is important because young people have historically been the tech industry’s most reliable early adopters. Losing this constituency is no small communication problem.
Calls to Slow Down AI Ring Hollow as Same Companies Accelerate
Smith had particular criticism of the increasingly common industry practice of calling for restraint while accelerating development. He particularly noted a recent article by Anthropic; This paper argued that it might be beneficial to slow down the advance of artificial intelligence “to give ourselves more time to deal with its enormous impacts.”
His response was clear: “If someone said, ‘This technology is so powerful we need a global agreement to slow it down,’ I would say: Well, if you think it’s moving too fast, take your foot off the gas pedal yourself.”
The debate comes down to an increasingly visible tension across the industry, where warnings about existential risk are often coupled with record levels of investment and product launches.
A Generation Shaped by Pandemic and Instability Deserves a Better Message
Smith placed the induction response in a broader biographical context, drawing attention to the particular circumstances of the group now transitioning to professional life. These are young people attending secondary school during the COVID-19 pandemic, whose social connections are primarily through screens, and who are coming of age in an environment of constant political fluctuations.
“Now they can finally join the workforce and AI is coming?” Smith said. “A lot of times it’s presented to them as something that is going to be theirs, not theirs. I think that’s their way of saying: ‘Wait a minute. Not so fast. We have a voice. We want to be heard.'”
Framing is important. Whether AI is something imposed on workers or something that expands what workers can do is not just a semantic question. It’s shaping how an entire generation approaches a technology that will define their professional lives.
The Real Timeline of AI Disruption is Decades, Not Quarters
One of Smith’s most salient objections to the dominant narrative concerns timing. He argues that the compressed forecasts commonly seen in investor presentations and the media are driven by fundraising mechanisms rather than evidence.
“This will emerge not in two and a half years, but in 25 years,” he said. “But look, if you’re trying to raise money, as entrepreneurs should, it’s easier to raise money if people think it’s going to happen sooner or later. But we spend a lot of time studying the history of technology. Twenty-five years for the complete transformation of an economy can be a record breaker.”
Compressing timelines serves capital formation. It does not serve workers, students, or anyone trying to make rational decisions about education and career.
Silicon Valley Keeps Making the Same Two Mistakes About Human Potential
At the heart of Smith’s critique is what he describes as a persistent double failure in how the tech industry thinks about its products and the people who will use them.
“Tech leaders tend to repeat two mistakes,” he said. “First: They overestimate the impact of technology, especially the speed it will achieve. Second: Tech leaders often underestimate humans. There was a time when humanity discovered that a horse could run faster than a human. So humans learned to ride horses. Let’s use AI to help humans do more and not replace us.”
The analogy is deliberately modest. The point is not that AI is unimportant, but that the pattern of human adaptation to powerful new tools has a long and consistent history that the current debate largely ignores.
Confident Predictions of Voices Proven to Be on the Wrong Foot Deserve Examination
Smith also took aim at the authority accorded to voices who gain credibility on unrealized predictions but continue to make sweeping statements with undiminished certainty.
“You’re seeing the same people who made wrong predictions a decade ago continue to make those predictions with extraordinary conviction,” Smith said. “And that’s great fodder for people who make stories for a living.”
AI Regulation Risks Replicating the Empty Stage of Social Media Policy Debates
On the question of governance, Smith drew a direct line between our current situation and the largely unproductive legislative debates surrounding social media over the past decade. According to him, this pattern repeats itself.
“You had some companies that said, ‘We want legislation,’ and then they opposed every bill in Congress because they never particularly liked it,” he said. He offered a practical filter for those steering AI policy discussions: Watch out for “ideas so grandiose that they have zero chance of adoption.”
Smith’s Main Argument: Redirect the Discussion to What Technology Can Build
There is one consistent theme in each of his criticisms. Conversations about AI and business are disconnected from the practical question of what the technology can actually do for the people who use it every day.
Smith said the discussion was “focused on too big a prediction” and not enough focus on: “Let’s use technology to help people do better things, like we’ve always done.”
Brad Smith has been at Microsoft for 33 years. His article on artificial intelligence and the future of employment was published by the company.
