March 2026 will be remembered as the month artificial intelligence crossed the threshold from experimental technology to operational necessity. In a span of just two weeks, a cascade of developments across enterprise, regulation, and national security has signaled that the AI era is no longer approaching—it has arrived. New data reveals that agentic AI—systems that can detect, decide, and execute tasks independently—has surged 31.5 percent to become the fastest-growing enterprise technology priority, with cybersecurity, sales, and supply chain management leading deployment . At the same time, PwC’s US chief executive has warned that partners who resist AI “have no place at the firm,” announcing a sweeping overhaul of the professional services business model . And in Washington, the White House has unveiled its National AI Legislative Framework, embracing a “try-first” mindset while House Financial Services Committee Chairman French Hill signaled support for regulatory sandboxes to enable responsible innovation . The message is unmistakable: the era of AI experimentation is over. The era of execution has begun.
The Agentic Revolution: From Copilots to Colleagues
The shift from generative AI to agentic AI represents a fundamental change in how enterprises think about automation. According to new research from The Futurum Group, which surveyed 830 global IT decision-makers, autonomous agents and agentic AI have surged 31.5 percent year-over-year to capture 17.1 percent of top-ranked technology investments—a dramatic acceleration from the second half of 2025 . When combining first- and second-place rankings, agentic AI now reaches 39.3 percent, up from 32.0 percent in late 2025.
“The pilot phase of enterprise AI is over,” said Keith Kirkpatrick, Vice President and Research Director at The Futurum Group. “Buyers have moved past prompt-based copilots and are now demanding AI that can detect, decide, and execute tasks independently. Vendors that continue to lead with generative AI assistants risk being outpaced by competitors who can demonstrate truly autonomous agents operating across production workflows” .
The deployment data reveals where this transformation is happening first. Cybersecurity leads planned agentic AI deployment at 58.7 percent, followed by sales, marketing, and service at 51.3 percent, and supply chain management at 47.8 percent . These are not experimental pilots—they are production-grade deployments targeting core business functions where autonomous systems can deliver measurable returns.
At the same time, the landscape of AI development is becoming increasingly competitive. Cursor, the San Francisco-based AI coding startup, announced plans to unveil Composer 2—an AI agent designed to carry out lengthy coding tasks on a user’s behalf—as it races to keep pace with larger rivals like Anthropic and OpenAI . With more than 1 million daily users and a reported $50 billion valuation, Cursor’s move underscores the intensifying battle to dominate the agentic AI market. Meanwhile, OpenAI released its most capable small models to date—GPT-5.4 mini and GPT-5.4 nano—with GPT-5.4 mini delivering more than double the speed of its predecessor while approaching the performance of much larger models .
The PwC Warning: ‘No One Gets a Free Pass’
Perhaps the most striking signal of AI’s transformative force came from PwC’s US chief executive, Paul Griggs, who declared in an interview with the Financial Times that partners who resist AI will have no place at the firm . The warning was part of a sweeping announcement that PwC is overhauling its services and pricing models to protect its business from being undermined by the technology.
“I don’t think anyone gets a free pass here. Anyone,” Griggs said. Anyone who believed they had the “opportunity to opt out” of AI is “not going to be here that long” .
The centerpiece of PwC’s strategy is “PwC One,” an AI platform that will offer clients access to six automated services—ranging from M&A due diligence to tax rules—on a subscription or consumption basis, potentially without a PwC person in the loop . The move represents a fundamental break from the traditional professional services model, which has long relied on billing clients by the hour for armies of junior staff performing routine tasks.
“Over time, it will move more and more of our work to outcomes pricing, which I believe our clients will readily accept because, ultimately, the only thing our clients care about is the outcome delivered,” Griggs said .
The hiring implications are already visible. “Am I recruiting the same number of accountants and traditional consultants vis-à-vis engineers, on a proportionate basis that I was three years ago? No,” Griggs acknowledged, adding that PwC is hiring more data specialists and engineers . While the firm remains “a net acquirer of talent at this point in time,” the composition of that talent is shifting decisively.
The Jobs Reality: AI Layoffs to Increase Ninefold—But Remain a Fraction of Doomsday Predictions
As AI capabilities expand, so does anxiety about job displacement. New research from the National Bureau of Economic Research, based on a survey of 750 chief financial officers from U.S. firms, offers a more nuanced picture than the dire predictions from some tech leaders .
According to the study, only 44 percent of surveyed CFOs say they plan some AI-related job cuts. When the co-authors calculated what that amounts to across the broader economy, they found just 0.4 percent—about 502,000 roles out of approximately 125 million—are expected to be lost this year . That represents a ninefold increase from the 55,000 AI-attributed layoffs in 2025, but remains a rounding error against the overall workforce.
“It’s not the doomsday job scenario that you might sometimes see in the headlines,” John Graham, co-author of the study and director of the Duke CFO survey, told Fortune .
The study also reveals a striking gap between perceived and actual productivity gains from AI. While tech leaders have made sweeping claims about AI-driven efficiency, Graham noted that “companies have invested and they’re realizing all these kind of cool things that they’re either starting to do or they hope to do in the near future. But it’s not really showing up yet in revenue” .
This lag echoes what economists call Solow’s paradox—the observation by Nobel Laureate Robert Solow in 1987 that “you can see the computer age everywhere but in the productivity statistics.” Goldman Sachs senior economist Ronnie Walker noted earlier this month that “we still do not find a meaningful relationship between productivity and AI adoption at the economy-wide level” .
Yet the study’s authors caution against complacency. “Who knows what’s going to happen in 2028?” Graham said. “I’m not making a prediction that there will never be any jobs lost two, three and five years from now to AI” .
The Washington Response: A ‘Try-First’ Mindset for AI Regulation
As AI reshapes the economy, policymakers are racing to establish frameworks that balance innovation with safeguards. On March 20, the Trump administration released its National AI Legislative Framework, embracing what House Financial Services Committee Chairman French Hill called a “try-first” mindset .
“I commend President Trump for releasing a thoughtful National Legislative Framework for AI that pairs innovation with targeted safeguards,” Hill said in a statement . “As Congress moves forward, advancing regulatory sandboxes and a flexible, sectoral based approach will be critical to enabling responsible development. The United States must maintain a ‘try-first’ mindset to preserve our leadership in the global AI race” .
The framework aligns with bipartisan efforts already underway in the House Financial Services Committee, which advanced a resolution on AI in financial services 54-0 in January 2026 . The resolution underscores the importance of appropriate oversight and robust consumer protections while affirming a pro-innovation approach.
The contrast with European approaches is stark. Arthur Mensch, co-founder and CEO of French AI company Mistral, has proposed a revenue-based levy on commercial AI providers operating in Europe, with proceeds flowing into a central fund to support content creation . “Europe does not need to choose between protecting its creators and competing in the AI race,” Mensch wrote in the Financial Times. “It needs a framework that enables both” .
The National Security Frontier: Palantir’s Maven Achieves ‘Program of Record’ Status
Perhaps the most consequential development of the month occurred at the intersection of AI and national security. The U.S. Defense Department will establish Palantir Technologies Inc.’s Maven AI system as a “program of record,” Reuters reported, a designation that smooths the system’s incorporation across all branches of the military as the Pentagon rapidly integrates AI into combat operations .
Maven, originally developed as Project Maven by the Pentagon’s Algorithmic Warfare Directorate, uses machine learning to analyze drone surveillance footage and has been credited with dramatically accelerating the military’s ability to identify targets. The program of record status represents a formal commitment to fielding the system across the force—a significant milestone for Palantir and a signal that AI is moving from experimental military applications to operational reality.
The Infrastructure Reality: Nvidia’s $26 Billion Bet and the Global Chip Crunch
Behind these developments lies a physical constraint that could shape the AI race for years to come. Nvidia disclosed in a securities filing that it plans to invest $26 billion over the next five years in developing open-source AI large language models—a move that dramatically expands the company’s ambitions beyond its traditional role as the leading hardware provider for AI workloads .
The investment, which dwarfs the $3 billion reportedly spent training GPT-4, will cover the full AI industry chain, with the first proprietary open-source models expected by late 2026 or early 2027 . The move signals that Nvidia is no longer content to be the “pick-and-shovel” provider in the AI gold rush—it is becoming a direct competitor in the model race.
Yet the supply side remains under pressure. Samsung’s largest union is threatening the company’s biggest-ever strike, with a vote scheduled to determine whether 5,000 workers will walk off the job in May, potentially disrupting chip production and exacerbating global semiconductor shortages . The strike threat underscores the fragility of the supply chains underpinning the AI revolution.
The Outlook: The Age of Execution
As March 2026 draws to a close, the contours of the AI era are becoming visible. Agentic AI is moving from pilot to production at unprecedented speed, with enterprises deploying autonomous systems across cybersecurity, sales, and supply chains. Traditional business models—including the billable hour that has defined professional services for generations—are being upended. Labor markets are beginning to adjust, with AI-related layoffs increasing even as overall employment effects remain modest. Policymakers in Washington are embracing a “try-first” approach to regulation, while the Pentagon formally integrates AI into combat operations. And the infrastructure powering it all—chips, models, talent—is straining against physical and geopolitical limits.
The common thread across these developments is that the era of asking “what can AI do?” is over. The question now is “how will we manage it?” The technology is no longer a future possibility but a present reality, embedded in enterprise operations, shaping policy debates, and restructuring labor markets.
As PwC’s Griggs put it: “Over time, it will move more and more of our work to outcomes pricing.” The same could be said of the broader AI transformation. The age of execution has arrived. The outcomes—economic, social, geopolitical—will define the decade to come.