The Morning Download: AI Boosts Productivity of European Firms

Good morning. AI already may be raising the productivity of European firms. A column published by the London-based non-profit Centre for Economic Policy Research used survey data to examine how AI adoption affects productivity and employment across more than 12,000 European firms. Highlights from the article that was published on Tuesday:

• The authors found that “AI adoption increased labour productivity levels by 4% on average in the EU, with no evidence of reduced employment in the short run.”
• The gains are not distributed evenly. “Medium and large firms, as well as firms that have the capacity to integrate AI through investments in intangible assets and human capital, experience substantially stronger productivity gains.”

The research underscored that to fully realize productivity gains from AI, leaders must make investments in a range of infrastructure, including software, data and worker training. The latter “amplifies AI’s productivity gains by 5.9 percentage points.”

The column adds to a body of research that captures the economic impact of AI. Stanford economist Erik Brynjolfsson, speaking last week at the WSJ Technology Council Summit, told the WSJ Leadership Institute’s Wendy Bounds that AI may be a factor in rising productivity.

“My estimate of productivity...is that…it's probably about 2.7% in 2025, which is roughly double what it was during the previous 10 years,” Brynjolfsson said. While other factors may be driving that rise in productivity, he determined that “AI-exposed occupations” accounted for a disproportionate share of lost jobs.

It took quite a few years for the full impact of the internet on productivity to reveal itself. AI’s impact is probably in the very early stages.

Accenture's adopt AI or you're out policy. As the tech consultancy asks clients to trust it as an AI transformation expert, it's pressuring its own senior staff to get on board. The FT reports that the firm is now tracking senior employees' weekly usage of its AI tools, requiring "regular adoption" for advancement to leadership roles.

The policy targets associate directors and senior managers, though staff in 12 European countries and its U.S. federal government division are exempt., FT reports.  Some employees have criticized Accenture's tools as ineffective, with one threatening to quit. 

Chief Executive Julie Sweet last year told investors that the consultancy had trained about 70% of its roughly 779,000 employees in generative AI fundamentals. But employees for whom “reskilling, based on our experience, is not a viable path” will be shown the door, Sweet said.

New report: Working (and working) with AI. A study of a 200-person tech company found that AI tools didn't reduce workload, but intensified it, according to Aruna Ranganathan and Xingqi Maggie Ye, both of the Haas School of Business, UC Berkeley.

Over eight months, employees worked faster, took on broader responsibilities and let work bleed into personal time all because AI made doing more feel effortless and rewarding. “But these experiments accumulated into a meaningful widening of job scope,” they write in Harvard Business Review. Over time these actions “produced a workday with fewer natural pauses and a more continuous involvement with work.”

Pro tip: The researchers suggested that companies adopt an “AI practice,” or routines that outline how AI is used, when it is appropriate to stop and the degree to which work expands with new capabilities.

Data center burnout. Trump loyalists in Red States are beginning to organize against the massive AI data center buildout, an agenda that has been pushed heavily by the White House and its allies, FT reports. “I voted for this administration and didn’t really think about [AI] until it started to affect me,” says one person, who lives near a Missouri site greenlit for a 400-acre data center.

The WSJ reported Wednesday how the buildout has triggered political opportunities for anti-data center candidates in places like Prince William County, Va., where AI infrastructure is starting to crowd out development of anything else. “Nothing can live next to data-center development like this except more data-center development,” Elena Schlossberg, a homeowner and activist, tells the Journal.

The buildout also threatens to strain local power grids, with residents and officials growing increasingly concerned that surging energy demand from data centers will drive up electricity bills for everyday consumers."

Illinois Gov. JB Pritzker is the latest politico to make it an issue, this week touting a two-year moratorium on new state-issued tax credits for data centers, in an effort to slow down power bills, Bloomberg reports.

Meta’s two-front battle prioritizing growth and innovation over safeguards. Mark Zuckerberg took the stand Wednesday in a personal-injury lawsuit filed by a woman who alleges that childhood social-media use led her to struggle with body dysmorphia, suicidal thoughts, anxiety, addiction and depression. The Journal reports that Meta and YouTube face some 3,000 lawsuits alleging liability for algorithmic features like infinite scroll and autoplay that make it difficult for teens to log off.

As the company fights claims, Meta is simultaneously fighting to keep AI unregulated. The New York Times reports that Meta is spending $65 million in 2026 to influence state elections in favor of AI-friendly politicians. The company has quietly launched four super PACs, split along party lines. The move reflects broader tech industry anxiety over a growing patchwork of state AI laws that companies fear could stifle innovation.

Meanwhile, the push to ban teens from social media goes global. From Paris to New Delhi, limits on children’s access to apps are now being debated or implemented, the Journal reports. In the U.S., Florida says it has started enforcing a ban on social-media use under age 14, and some states, including California and New York, have passed legislation requiring warning labels.

Highlights from this week's global AI summit in New Delhi.

• Tata Consultancy Services is teaming up with OpenAI to develop industry-specific AI services and build AI infrastructure in India, the WSJ reports. Rival Infosys announced a similar tie-up with Anthropic days earlier.
• India's Reliance Industries announced plans to invest $110 billion to build large-scale data centers in the country and offer AI services.
• OpenAI and Anthropic's rivalry was on display during an event with Indian Prime Minister Narendra Modi and other major tech figures, CNBC reports.

AI pioneer’s startup raises $1 billion. World Labs, the AI startup founded by pioneer Fei-Fei Li, raised $1 billion in new funding, Bloomberg reports. The company focuses on "world models," AI designed to navigate and make decisions in three-dimensional environments.

The U.S. is sending significant numbers of jet fighters and support aircraft to the Middle East, assembling the greatest amount of air power in the region since the 2003 invasion of Iraq. (WSJ)

British public broadcaster the BBC said police had arrested Andrew Mountbatten-Windsor, the former Prince Andrew, on suspicion of misconduct in public office. (WSJ)

A Seoul court sentenced former South Korean President Yoon Suk Yeol to life in prison for leading an insurrection during a short-lived attempt to impose martial law. (WSJ)

WSJ Tech Live Cybersecurity brings together leading voices from government, enterprise technology, venture capital and cybersecurity for a full day of newsroom-led conversations. Join us in the heart of New York City as we examine the evolving threat landscape, the role of AI, policy priorities and how organizations are responding to increasingly complex cyber risks.

Request your invitation with code WSJPRO to save 20% on your pass. Please note that all requests are carefully reviewed, and passes are limited, so we encourage you to reserve yours soon.

The WSJ CIO Journal Team is Steven Rosenbush, Isabelle Bousquette and Belle Lin.

The editor, Tom Loftus, can be reached at thomas.loftus@wsj.com.