The Department of Labor and AI Data Sharing

The Department of Labor (DOL) is working to secure data-sharing agreements with technology and Fortune 500 companies to help the Bureau of Labor Statistics assess how artificial intelligence is being implemented in workplaces and whether AI is displacing workers or changing job roles. This is consistent with the Administration’s AI Action Plan and voluntary rather than mandatory efforts.

• The Administration has made accelerating AI innovation and adoption a priority, and the Department’s proposal follows the release of an AI Action plan that calls for Federal agencies to collaborate to study the impact of AI on the labor market.
• The proposal also comes as Congress is considering several proposals -- some voluntary and others mandatory -- for collecting more data on the workforce impacts of AI.
• Participation in any data-sharing initiative would require coordination across teams including legal, HR, and technology. Companies will need to consider how they would avoid compromising trade secrets, personnel information, or other confidential data.
• Firms should ensure they have internal visibility into where and how AI tools are being used within their organizations to be prepared to respond to shifting Federal disclosure requirements.  

AI Innovation and Adoption is a Federal Priority

The Administration has made accelerating AI innovation and adoption a priority. The President’s January 2025 Executive Order directed the development of an AI Action Plan to sustain and enhance US AI leadership. The subsequent Plan outlined 90 policy changes needed to ensure the US wins the AI race and several principles for AI policy, including ensuring that US workers benefit from AI innovation. The Plan also recommended that BLS, Census Bureau, and Bureau of Economic Analysis collaborate to study AI’s impact on the labor market and that DOL establish an AI Workforce Research Hub to evaluate the impact of AI on the labor market. Earlier this year, the Department released an AI Literacy Framework for workforce and education systems and launched an AI in Registered Apprenticeship Innovation Portal to help employers and other organizations integrate AI skills into apprenticeship programs. 

The DOL initiative, announced at a recent conference, comes as policymakers, business leaders, and educators alike are closely tracking workplace AI adoption, particularly implications for employer skills requirements and overall employment trends.¹  However, data limitations make this a challenging question to study.  US statistical agencies also face broader pressure to modernize economic data collection. Response rates for voluntary DOL surveys, such as the jobs report and other economic indicators, for example, have steadily declined. At the same time, policymakers and researchers increasingly want more timely, detailed, and higher-frequency information.

Legislative Initiatives 

Congress is considering a range of proposals – some voluntary and others mandatory – for collecting more data on AI workforce impacts. A bipartisan bill, the Workforce Transparency Act, would direct DOL to collect and publish data provided voluntarily by public and private employers about the use of AI in the workplace. Congress is also considering more prescriptive proposals that would require disclosure of AI-related job losses. Last Fall, for example, a bipartisan pair of senators, including one of the Workforce Transparency Act’s sponsors, introduced the AI-Related Job Impacts Clarity Act (S. 3108), which would require covered entities to report AI-related job impacts, including layoffs, hiring, unfilled positions, and retraining substantially due to AI. DOL would then publish quarterly reports and underlying data. 

Potential Limitations 

A voluntary, company-based data collection model could provide important insights into AI labor trends. However, it could also suffer from selection bias if participating companies are not representative of the broader economy or if firms experiencing more sensitive workforce impacts decline to share data. In addition, even with better data, determining whether AI was the primary cause, a contributing factor, or merely part of a broader reorganization will remain difficult. 

Business Impacts

Employers should treat DOL’s emerging data effort as an early signal of a broader policy trajectory. Even if the Department begins with voluntary data-sharing partnerships, the questions Federal agencies and lawmakers are asking could shape future disclosure expectations, workforce funding priorities, apprenticeship models, and public benchmarks for responsible AI adoption. 

Companies deploying AI should build internal visibility into where AI tools are used, which tasks they support or replace, which employee groups are affected, and how AI use differs across roles, worksites, and business units. This does not require every organization to build a Federal-reporting system immediately, but it does suggest basic AI-use inventories, clear ownership across business functions, and documentation that connects AI adoption to workforce planning. 

Data architecture will be equally important. Participation in any DOL data-sharing initiative would require coordination among legal, compliance, HR, technology, privacy, and business teams. Companies should consider whether they can aggregate and anonymize relevant data, protect trade secrets and customer information, avoid disclosing individual-level personnel or performance information, and explain the limits of their data. These steps will matter whether information is shared voluntarily with Federal agencies, summarized for investors, or used internally for board-level oversight. 

Better Federal data on AI adoption could influence workforce grants, apprenticeship standards, training priorities, and expectations around reskilling. Employers should monitor how DOL, BLS, and Congress define AI use, job impacts, and skill needs, because those definitions may shape future policy and may also influence how external stakeholders assess corporate AI adoption. 

Finally, businesses should improve documentation around restructuring and role redesign. When AI is cited as a factor in staffing decisions, companies should be able to distinguish productivity improvements, task changes, reductions in force, hiring shifts, and retraining commitments. Clear records will help leaders explain decisions internally and externally while reducing the risk that incomplete data or inconsistent terminology create compliance, reputational, or employee-relations challenges.