The Workforce Transparency Boom Is Fading
For nearly 60 years, the federal government has collected workforce demographic data from tens of thousands of American employers through the EEO-1 reporting system.
That system may now be ending.
On May 14, the EEOC submitted a proposal to the White House’s Office of Information and Regulatory Affairs (OIRA) to rescind EEO-1 collection and reporting entirely.
The proposal is broader than it first appears. It would also eliminate demographic data collection covering union membership, state and local government workforces, and public school employees—EEO-2, EEO-3, EEO-4, and EEO-5 collections.
The move immediately drew criticism from former EEOC officials, along with a bipartisan response from lawmakers led by Rep. Jerry Nadler and Sen. Lisa Blunt Rochester calling on the agency to continue collecting the data.
Importantly, the proposal must still clear a multi-step regulatory process—including OIRA review, a Federal Register comment period, and formal rulemaking—before taking effect. That timeline is tight relative to the September 30 filing deadline, meaning the 2025 collection cycle could still proceed. As of this writing, however, the filing portal has not opened; last year it launched on May 20.
The EEOC was created in 1965. EEO-1 collection began in 1966, initially covering roughly 25 million employees at 45,000 employers. Today, the framework covers approximately 56 million employees across 73,000 employers.
At the same time, discrimination complaints filed with the agency remain near historical highs. The EEOC received 88,201 charges in FY2025—not far below the roughly 100,000 annual filings seen at the peak in 2010–2012, and well above the pandemic-era low of 61,331 in FY2021.
Which makes another detail particularly striking: on the very same day the EEOC proposed ending future EEO-1 collection, the agency also announced a sex discrimination settlement that relied on the employer’s own historical EEO-1 filings.
For decades, however, the public only had access to aggregate statistics published by the EEOC. Outside of a small group of mostly technology companies that began voluntarily disclosing around 2013–2014, company-level workforce demographic data was largely unavailable.
That changed after 2020.
Following investor pressure and corporate commitments after George Floyd’s murder, public disclosure of EEO-1 workforce data became mainstream among large U.S. companies. Nearly 80% of the current S&P 500 disclosed workforce demographic data for 2023 or 2024, while disclosure among the S&P 100 exceeded 90%.
The trend has now reversed.
S&P 500 disclosure has fallen to below 60%, while Russell 1000 disclosure declined to under 40%. The pullback has been especially pronounced among smaller public companies, where disclosure never became widespread to begin with.
The retreat is not entirely surprising. Many of the investors and proxy advisors that originally pushed companies toward workforce transparency—including BlackRock, State Street, ISS, and Glass Lewis—have significantly softened or scaled back their public expectations around diversity disclosure over the last two years.
Even if the federal government ultimately ends EEO-1 collection, the underlying data may not disappear entirely. California, Massachusetts, and Illinois maintain state-level reporting requirements for certain employers, and Illinois law currently makes all filings publicly available.
Still, the broader direction is clear: fewer quantitative disclosures, less standardization, and growing uncertainty around one of the most important workforce datasets in corporate America.
At PeopleReturn, we track workforce disclosure across the Russell 3000 and maintain the largest database of public-company EEO-1 data assembled from company disclosures, government databases, and historical filings.
Get PeopleReturn insights in your inbox
“Lower-Value Human Capital”
On May 19, Standard Chartered CEO Bill Winters told reporters at a Hong Kong investor event that the bank’s plan to eliminate nearly 8,000 support roles over the next four years was not really about cost cutting.
“It’s replacing in some cases lower-value human capital with the financial capital and the investment capital we’re putting in,” he said.
The phrase spread almost immediately across social media in Asia. Former Singapore president Halimah Yacob publicly rebuked the remarks, writing that workers facing retrenchment “are human beings with families, not just a form of capital.” Regulators in both Hong Kong and Singapore reportedly contacted the bank for clarification. Winters later sent an internal memo to staff, and the bank said his comments had been taken out of context.
The plan itself, however, did not change. The jobs are still being cut.
The episode matters less for what it says about Standard Chartered specifically than for what it accidentally made explicit.
“Lower-value human capital” is increasingly the underlying logic of AI-era workforce restructuring. Executives may use softer language publicly, but the broader calculation is becoming familiar across corporate America: if AI systems can perform certain functions faster, cheaper, or at greater scale, companies will eventually reduce the labor supporting those functions.
Saying that publicly was the mistake. The restructuring itself is becoming normalized.
That same week, Intuit announced plans to eliminate roughly 1,800 roles while continuing to aggressively expand its AI investments. CEO Sasan Goodarzi linked the restructuring directly to surging internal AI adoption and what he described as an “agentic AI-first operating model.” Investors were unconvinced; shares fell sharply following the announcement.
Similar patterns are emerging across industries. Novo Nordisk and Nestlé have both recently announced major workforce reductions while emphasizing a new “performance culture” focused on faster decision-making and leaner organizational structures. Microsoft, Block, Shopify, Coinbase, and other large employers have likewise linked workforce reductions, flatter organizations, or higher performance expectations to AI-driven productivity gains.
What is changing is not merely the scale of restructuring, but the framing around it.
For years, corporate discussions around layoffs were generally positioned around cyclical slowdowns, macroeconomic conditions, or isolated restructurings. Increasingly, however, labor is being discussed more explicitly as one input among many—one now being evaluated directly against software, automation, and AI agents on cost and productivity grounds.
That framing may be analytically coherent. As Winters discovered in roughly 12 hours, it does not survive contact with the people it describes.
For investors and companies focused on human capital, this creates a genuine tension. The productivity gains from AI are real and increasingly reflected in market valuations. But the workforce implications — who is being displaced, from which functions, and in which geographies—are becoming harder to assess precisely as workforce transparency begins to recede.
The irony is that companies are restructuring around workforce productivity and organizational efficiency at exactly the same moment public visibility into workforce composition, sentiment, and employee experience may be declining.
Meta’s AI Transition Is Showing Up in Employee Sentiment
Meta spent the past two years convincing investors that AI would make the company faster, leaner, and dramatically more profitable.
So far, Wall Street seems convinced.
Meta is generating enormous profits, ramping capital expenditures toward as much as $145 billion this year, reorganizing teams around AI, transferring thousands of employees into AI-focused roles, and laying off roughly 8,000 workers globally.
Inside the company, however, the mood appears considerably darker.
Recent reporting from the New York Times and WIRED describes historically low morale, widespread anxiety about layoffs, frustration over mandatory AI adoption, and employee backlash against new internal tracking systems designed to help train AI models. Some employees reportedly believe they are helping build systems that will ultimately reduce the need for workers like themselves.
That dynamic is increasingly showing up in the data.
Using our proprietary recency-weighted sentiment model built on employee reviews, Meta’s modeled employee rating has fallen from just over 4.0 in early 2025 to roughly 3.71 today—a meaningful decline for a company that historically maintained unusually strong employee sentiment.
The drop accelerated almost precisely as Meta intensified its AI restructuring: large-scale layoffs, forced internal transfers, productivity mandates, workforce monitoring initiatives, and increasingly explicit messaging around “efficiency.”
The more important question is whether this is primarily a Meta story — or an early indicator of something broader.
Meta is hardly alone. Executives at firms including Block, Microsoft, Coinbase, Shopify, and Intuit have all framed workforce reductions around AI-driven efficiency gains. For years, investors evaluated AI primarily through infrastructure spending, productivity gains, margin expansion, and competitive positioning. The human capital dimension—how employees actually experience these transitions—received comparatively little attention.
That is changing.
AI adoption does not just restructure workflows. It reshapes how employees perceive job security, fairness, autonomy, and the basic relationship between labor and management. Those perceptions affect retention, productivity, and organizational trust in ways that do not show up immediately in earnings but tend to show up eventually.
AI strategy, in other words, is rapidly becoming workforce strategy.
And the effects are becoming measurable in real time—which is precisely why workforce transparency may become more important just as many companies disclose less.
What we’re reading:
- The Rise of Emotional Surveillance Every successful technology needs to find a problem that people are willing to pay money to solve. In the case of emotion AI, that problem appears largely, so far, to be worker performance and productivity, especially in customer service and blue-collar labor. (The Atlantic)
- Investors putting US boards on notice over shareholder proposal omissions Some of the US’s largest public pension funds are voting against board members at companies in the country over shareholder proposal omissions this proxy season. (Responsible Investor)
- Forget the rockets. NASA’s real achievement is its ‘lifers’ In an era when employee retention dominates the HR agenda, NASA’s numbers are remarkable. Traster estimates (NASA) Glenn’s annual turnover rate sits somewhere between 2 and 5%, and the majority of that is retirements. Voluntary departures are rare enough to barely register. (HRD)
- These Companies Say AI Is Reviving Entry-Level Jobs, Not Killing Them Employers have singled out AI in recent years as a reason to cut back on hiring college graduates. Now some companies are flipping that narrative, saying AI is boosting their need for entry-level workers. (Wall Street Journal)
- Court Order Signals New Era for Shareholder Proposals Under Rule 14a-8 A recent federal court decision compelling a public company to include a shareholder proposal in its proxy materials illustrates the importance of shareholder engagement following changes to the Securities and Exchange Commission’s administration of Rule 14a-8. (Debevoise & Plimpton)
- DEI Isn’t Dead. But It’s Not Really Alive, Either Diversity, equity and inclusion efforts might look different than they used to, and managers might be talking about them in a new way, if they’re talking about them at all.1But they’re still there — even if companies have abandoned the now taboo three-letter acronym. (Bloomberg) and DEI experts say the acronym may be radioactive, but the underlying business case is stronger than ever (Fortune)
- This data helped fight job discrimination. Trump wants to get rid of it Just how diverse is corporate America? For decades, it has been the federal government’s job to keep track. Not for much longer. The Trump administration has signaled it’s putting a stop to it as part of its diversity, equity and inclusion reforms. (USA Today)
- A Nobel economist figured out 60 years ago that people learn best on the job. AI is making that almost impossible Sixty years ago, an economist named Kenneth Arrow sat down and worked out something that seemed almost too obvious to say: workers get better at their jobs by doing them. Now, as artificial intelligence chips away at the entry-level jobs that once served as the on-ramp to white-collar careers, researchers at the Federal Reserve Bank of Atlanta are dusting off Arrow’s 1962 paper and warning that companies racing to automate their way to lower payroll costs may be sawing off the branch they’re sitting on. (Fortune)
- CEOs Are Getting Ruthless About Worker Performance As the looming threat of artificial intelligence gives employers more leverage in an already sluggish white-collar job market, a growing chorus of CEOs in every major sector is saying the quiet part out loud, and frequently underscoring the point with layoffs. (Bloomberg)
- Return-to-Office Mandates are Triggering a New Era of Workplace Sabotage The corporate world is currently locked in a high-stakes psychological standoff. While CEOs champion the office as a hub for “serendipitous innovation” and “culture-building,” the view from the cubicle looks radically different. New data suggests that the push to bring employees back to the desk has triggered a profound breach of the employer-employee social contract. (Enhancv)