Artificial intelligence (AI) is no longer just the product Silicon Valley is selling. It’s increasingly the reason companies are rethinking how many people they need and what kind.
Across Big Tech and IT services, AI is showing up in earnings calls, restructuring plans, hiring memos and layoff notices. The shift isn’t uniform, and it isn’t always framed as “AI replacing jobs.” But read between the lines, and a pattern is emerging where companies have leaner teams, slower hiring, and a sharper divide between routine roles and high-skill AI work.
The capability is already here
In November 2025, researchers at Massachusetts Institute of Technology released the Iceberg Index, a skills-based model simulating 151 million US workers. It found that AI can technically perform work equivalent to 11.7% of US wage value, which is roughly $1.2 trillion.
The visible impact so far has been concentrated in computing and tech (about 2.2% of wage value, approximately $211 billion), but the larger exposure sits beneath the surface, in administrative, financial and professional services work. The study measures technical overlap — not job losses — but it underscores how much white-collar work is now within AI’s reach.
Inside tech companies, that overlap is starting to translate into operational change.
“Do more with fewer people”
At Amazon, CEO Andy Jassy warned in June 2025 that AI agents would likely reduce corporate roles over time by automating routine tasks.
“As we roll out more Generative AI and agents, it should change the way our work is done. We will need fewer people doing some of the jobs that are being done today, and more people doing other types of jobs. It’s hard to know exactly where this nets out over time, but in the next few years, we expect that this will reduce our total corporate workforce as we get efficiency gains from using AI extensively across the company,” he noted in a company blog post.
In October, the company confirmed plans to cut about 14,000 corporate jobs, with more reductions expected next year. However, Jassy attributed the move to “culture” and not AI.
“The announcement that we made a few days ago was not really financially driven, and it’s not even really AI-driven, not right now at least,” he said in the company’s earnings call at the time. “It really — it’s culture.”
At the same time, Reuters reported that Amazon was expected to have spent roughly $118 billion in capital expenditure in 2025, much of which is tied to AI and cloud infrastructure.
At Salesforce, CEO Marc Benioff has been unusually blunt. He said in a podcast interview with investor Logan Bartlett that internal AI use allowed him to reduce customer support headcount from 9,000 to about 5,000. Earlier, he told Bloomberg that AI was already doing as much as 50% of the work at the company. Salesforce has rolled out its AI product, Agentforce, while trimming roles and redeploying some staff.
In early 2025, Microsoft CEO Satya Nadella, while speaking at Meta’s inaugural LlamaCon AI developer event, said that about 30% of the company’s code is now written by AI.
Speaking with Nadella, at the same event, Meta CEO Mark Zuckerberg added, “Our bet is sort of that in the next year probably … maybe half the development is going to be done by AI, as opposed to people, and then that will just kind of increase from there.”
At IBM, CEO Arvind Krishna told The Wall Street Journal that AI agents replaced the work of a few hundred HR employees. But he also said IBM’s overall headcount has risen, as savings were redirected into hiring programmers, salespeople and AI specialists. Speaking with Bloomberg in 2023, Krishna said, “I could easily see 30% of that getting replaced by AI and automation over a five-year period.”
The messaging varies — some cuts are framed as culture resets, others as efficiency drives — but AI is increasingly part of the calculus.
AI-first as company policy
Some companies aren’t just adopting AI; they’re restructuring around it.
In early 2025, Shopify CEO Tobi Lutke told staff that before asking for more headcount, teams must demonstrate why AI cannot do the job. AI usage is now part of performance reviews.
Around the same time, Duolingo CEO Luis von Ahn declared the company “AI-first,” saying headcount will only be approved if automation isn’t possible and that contractors will gradually be phased out where AI can handle the work.
Freelance services marketplace Fiverr cut roughly 30% of its workforce as it pivots to what CEO Micha Kaufman called an “AI-first company.” HP plans to cut between 4,000 and 6,000 roles by 2028, citing AI-driven efficiencies among other cost measures. Meta has trimmed roles within its AI unit to streamline decision-making even as it doubles down on infrastructure spending.
The throughline is that companies want AI literacy baked into every function. In some cases, that’s about augmentation. In others, it’s about replacing repetitive work and compressing teams.
The IT services inflection point
Nowhere is the shift more visible than in IT services.
Tata Consultancy Services cut more than 12,000 jobs in 2025, largely at mid and senior levels, citing skill mismatches. Industry experts told Reuters that up to 400,000-500,000 roles across India’s $283 billion IT sector could be at risk over the next two to three years as AI handles coding, testing and support functions.
India’s top five IT firms added just 17 net employees in the first nine months of FY2025-26, compared with 17,764 in the same period a year earlier, according to The Times of India. The report noted that AI-led delivery models are decoupling revenue growth from headcount growth, which is a structural shift in a sector that historically scaled by hiring more engineers.
Firms like Accenture and SAP are restructuring too but with a heavy emphasis on reskilling and changing the “skill mix.” The idea isn’t necessarily to shrink permanently, but to phase out nonviable skills and redirect investment toward AI-heavy growth areas.
Slower hiring, sharper expectations
Even companies building AI are adjusting growth plans.
Just last month, CEO Sam Altman said OpenAI would “dramatically slow down” headcount growth because AI enables the company to do more with fewer people. The company isn’t freezing hiring, he said, but wants to avoid overexpansion followed by painful corrections.
A December 2025 report from Microsoft Research, based on 200,000 Copilot conversations, found high AI applicability in computer and mathematical occupations, including data scientists and developers.
Still, entry-level roles appear vulnerable. Stanford economist Erik Brynjolfsson told the Washington Post that demand for some junior roles has fallen, while demand for senior AI expertise is soaring.
Replacement, augmentation or compression?
The tech sector isn’t witnessing a single, clean story of “AI takes jobs.” Instead, it’s seeing compression.
Fewer layers of management. Fewer routine back-office roles. Slower hiring at the junior level. Higher expectations that every employee knows how to use AI tools.
The MIT Iceberg Index suggests the technical capacity for automation is significant. But adoption decisions, reskilling efforts and economic conditions will determine how that capability translates into real job losses.
For now, growth in tech no longer automatically means headcount growth. In the AI era, companies are trying to prove they can scale revenue, output and products without scaling payroll at the same rate.
And that may be the most disruptive shift of all.