ビッグテックCEO、AI失業論から転換 雇用創出を主張

Throughout 2024, executives at major U.S. technology firms sounded alarms about artificial intelligence's threat to employment. OpenAI's Sam Altman and Anthropic's Dario Amodei were among those warning that AI could displace workers at scale. By early 2025, however, that narrative had shifted markedly.

In late May, Altman recalibrated, saying the industry had "underestimated how much humans can remain central in everything." Amodei, who had warned in May of the prior year that AI might eliminate half of entry-level jobs, now frames AI adoption differently: companies using AI can "accomplish more with the same resources." Meta's Mark Zuckerberg made an explicit claim that if companies increase human productivity faster than they automate, "future employment should theoretically increase rather than decrease." Amazon's Andy Jassy, who in February mentioned AI's job-creation potential, had previously announced workforce reductions and in 2024 cut 16,000 employees—a move he attributed not to AI but to organizational restructuring and culture initiatives.

The credibility gap is stark: Meta laid off 8,000 workers in May while its CEO spoke of employment growth; Amazon cut 16,000 while positioning AI as job-positive. This contradiction has bred skepticism that tech leaders are primarily seeking to rehabilitate AI's public image rather than reporting genuine labor market shifts.

Yet broader business sentiment suggests the shift may reflect real observation. EY-Parthenon's quarterly surveys show a dramatic move in CEO opinion: the proportion expecting major workforce reductions from AI investment fell from roughly 46% in January 2025 to 20% in May 2025. A joint study by fintech company Lamp and workforce intelligence firm Revelio Labs adds early empirical support: firms most actively investing in AI are hiring approximately 10% more employees than comparable companies that have not yet adopted AI. This hiring differential, if sustained, would indicate that productivity gains from AI are creating new roles faster than automation eliminates existing ones—lending credence to the tech leaders' revised claims, even as questions about longer-term labor displacement remain unresolved.

The reversal in messaging from tech leaders reflects a calculated recalibration as AI deployment moves from research to commercial operation. Last year's warnings about job destruction were based on theoretical worst-case scenarios; now that companies are actually implementing AI systems, executives appear to be observing that productivity gains can offset automation—a finding supported by EY-Parthenon's sharp drop in CEO pessimism over just four months. This shift is not purely coincidental or benign: tech leaders have obvious incentives to downplay job losses (regulatory scrutiny, talent recruitment, brand reputation), and the fact that their own companies continue to cut staff even while making pro-employment claims invites reasonable skepticism.

However, the emerging data from Lamp and Revelio Labs suggests the optimism may not be entirely rhetorical. Companies that have adopted AI most aggressively are hiring roughly 10% more employees than non-adopters, which could indicate that AI-driven efficiency truly is creating net new headcount rather than simply automating existing roles away. The timing matters: these are early-stage findings, and the full labor market impact of generative AI remains uncertain. The narrowing gap between tech-leader statements and measurable outcomes over the next 12–24 months will likely determine whether the pessimism of 2024 was premature or whether current optimism is premature.

No comments yet. Be the first to share your thoughts!