The clearest case is TTEC, a customer experience tech firm. In an April 30 memo obtained by Business Insider, Chief People Officer Laura Butler told employees the company would suspend discretionary 401(k) matching contributions starting in the second quarter of 2026, with the pause expected to run through year-end.

Butler described the move as a “difficult decision” intended to create financial flexibility for “tools, training and capabilities” tied to the company’s future strategy, including AI tools, certifications and automation.

“The TTEC spokesperson confirmed to Business Insider that the company had temporarily suspended the discretionary 401(k) company match, saying the move was ‘part of a broader set of actions to create the financial flexibility needed to accelerate our business transformation,’” according to the publication’s coverage.

Read more: Deloitte and Zoom cut benefits; who’s next?

Deloitte, Zoom announce benefit cuts

Business Insider also reported in mid-April that Deloitte U.S. would reduce benefits for employees in its “Center” talent model, which covers internal support roles including IT, administration and finance. According to spokesperson comments to the publication, the firm plans to trim paid parental leave, scale back PTO, change pension arrangements and cut funding for in vitro fertilization benefits for that employee group.

Zoom also reduced the length of paid parental leave for both birthing and non-birthing parents compared with prior policies, Business Insider reported.

TTEC explicitly linked the 401(k) cut to AI‑related investments, while Deloitte and Zoom have not publicly framed their benefit cuts in the same way.

When benefits are cut, HR feels the impact

The risk to organizations is not just the individual policy change, but what employees make of the benefits-cutting pattern. A 2021 Washington State University–led meta‑analysis found that cost‑cutting actions such as layoffs and benefit reductions depress job satisfaction and organizational loyalty for at least two years, unless employers deliberately invest in remaining workers to signal that they remain valued.

However, when a prominent tech company trims benefits, the move may give other employers cover to follow in those footsteps. Additionally, workers watching these announcements may conclude that companies are willing to fund transformation by dialing back the benefits used to recruit them in the first place.

Although HR leaders cannot control every financial decision made above them, they can shape how those decisions are explained, implemented and received.

Communication at risk

Communication often becomes as important as the policy itself. If a benefit change can be connected even indirectly to AI spending or restructuring, leaders need to be specific about the rationale, the timeline and any conditions for restoration. Vague references to realignment or restructuring fill in a narrative that may actually be more negative than the reality.

Gallup research suggests that cutting or trimming benefits can lower short‑term costs, but often increases long‑term turnover and re‑hiring expenses if not offset by clear communication and alternative support.

Credibility can slide

There is also a longer-term credibility question. If organizations ask employees to accept more lean benefits now in exchange for technology investment and efficiency gains, they will eventually need to demonstrate that those gains translate into something for the workforce. Without a credible answer, engagement and retention problems can persist long after the immediate budget pressure lifts.

The post Deloitte, Zoom and TTEC benefits cuts highlight growing HR challenges appeared first on HR Executive.

“Over the past several years, more than one million businesses have adopted OpenAI’s products and APIs,” says a release from OpenAI. “Across those deployments, one pattern has become increasingly clear: the next stage of enterprise AI will be defined by how effectively businesses can deploy this technology into real-world use cases, and how well we and our Alliance partner ecosystem can help them.”

In February, HR Executive reported on OpenAI’s Frontier Alliances, a series of multi-year partnerships with McKinsey & Co., Boston Consulting Group, Accenture and Capgemini to help enterprises deploy Frontier, OpenAI’s platform for building and managing AI agents across company-wide operations.

Those consultants are hired to redesign operating models, manage workforce change and embed AI into day-to-day workflows. Deployment Company is a harder push in the same direction, which often crosses into HR territory.

What is OpenAI’s Deployment Company?

The new company will place Forward Deployed Engineers (FDEs) inside client organizations to work closely with business leaders, operators and frontline teams. Additionally, OpenAI has agreed to acquire Tomoro, an AI consulting and engineering firm, in a move that brings along approximately 150 specialists.

“Today’s launch of the OpenAI Deployment Company will help bridge a critical gap for customers,” wrote Sarah Friar, CFO at OpenAI, in a LinkedIn post. “It’s also the key reason we’ve agreed to acquire Tomoro … With a proven track record deploying AI in complex enterprise environments for companies like Tesco and Virgin Atlantic, the team will expand our ability to embed frontier AI engineers directly within organizations around the world.”

As HR Executive noted in February, OpenAI named no CHROs as stakeholders in the Frontier Alliances announcement. The announcement made no mention of people functions as co-owners of the transformation these firms would lead.

The Deployment Company launch follows the same pattern. FDEs work directly with business leaders and frontline teams to redesign organizational infrastructure and critical workflows around AI.

In practice, the engineers touch HR’s scope, including hiring pipelines, performance management systems and workforce planning tools. “We’re now well past experimentation,” wrote Friar. “The challenge in front of us is getting AI embedded deeply into the workflows that power businesses, with the engineering, operational rigor and change management required to make it stick.”

Anthropic in the enterprise

In related news, Anthropic recently announced a new enterprise AI services company backed by approximately $1.5 billion from partners, including Blackstone, Hellman & Friedman and Goldman Sachs, with contributions from General Atlantic, Leonard Green, Apollo Global, Singapore’s wealth fund IC and Sequoia Capital.

The venture deploys Anthropic’s applied AI engineers to embed Claude into mid-sized companies’ core operations. “Enterprise demand for Claude is significantly outpacing any single delivery model. Our partnerships with the world’s leading systems integrators are central to how Claude reaches large enterprises,” said Krishna Rao, chief financial officer of Anthropic, in a release. “This new firm brings additional operating capability to the ecosystem and capital from leading alternative asset managers.”

The post OpenAI’s new $4B consulting venture lands squarely in HR territory appeared first on HR Executive.

While AI has enabled efficiency, faster screening, broader reach and better matching, it has also given rise to massive candidate fraud amid a flood of mass-generated resumes. AI in hiring has quickly become a risk management challenge with serious legal, reputational and operational implications.

As organizations adopt AI tools to manage the unprecedented volume of applicants, they are discovering a sobering reality: Automation has lowered barriers for those intent on sophisticated fraud.

A recent survey of senior talent acquisition executives conducted by the Institute for Corporate Productivity (i4cp) found:

• 75% reported they have encountered candidate fraud in hiring in the past 12 months; 12% suspect this has happened but haven’t confirmed.
• 67% have detected fabricated credentials (diplomas, certificates, government IDs that can pass initial document screening).
• 40% have discovered location deception (remote candidates claiming to be in the U.S., using VPNs or IP spoofing to get higher pay while being located overseas).
• 40% have found fabricated digital footprints (e.g., fake LinkedIn profiles to accompany the resume, featuring AI-generated headshots, fabricated endorsements to pass background checks).

See also: Workplace lying is at crisis levels; time for deception detection?

For senior HR executives, the issue is no longer whether to use AI in hiring, but how to utilize it responsibly amid escalating candidate fraud and a rapidly shifting regulatory landscape. To further complicate things for global companies, AI hiring risk looks fundamentally different outside the United States—particularly in the European Union.

Under the EU Artificial Intelligence Act, AI systems used for recruitment, candidate screening, interviewing, performance evaluation, or termination decisions are explicitly classified as high‑risk systems. This designation triggers mandatory requirements that go far beyond current U.S. norms, including formal risk management programs, documented impact assessments, bias testing, human oversight, audit‑ready technical documentation and ongoing monitoring.

The new hiring reality: volume, velocity and vulnerability

Employers are drowning in resumes, many of them AI‑generated. Recruiters report being overwhelmed, while qualified applicants struggle to stand out. In response, companies have leaned even more heavily on AI‑enabled screening and assessment tools, often without fully accounting for the risks that accompany them.

More troubling still, the i4cp survey found that many employers acknowledge they have not established structure or governance around the use of AI in hiring. Most (56%) indicated that their organizations are in the “awareness/ad hoc” phase of AI risk management maturity.

For HR leaders, this convergence of applicant volume, sophisticated deception and lack of governance creates real exposure—not only to hiring the wrong person, but to legal liability and cybersecurity threats.

Fraud isn’t just a talent problem, it’s an enterprise risk

Fraudulent candidates introduce risks that extend far beyond poor performance. Negligent hiring claims remain one of the most significant legal exposures for employers, particularly when background checks are inadequate or inconsistent. Courts have repeatedly held employers liable when they “knew or should have known” that an employee posed a risk at the time of hire—an increasingly easy standard to meet now that fraud risks are well documented.

The implications escalate further when fraudulent hires gain access to sensitive systems or data. Law enforcement has warned that fake IT workers—sometimes tied to foreign criminal or state‑sponsored actors—have infiltrated U.S. companies to steal intellectual property, exfiltrate customer data, or deploy ransomware. In regulated industries, liability may extend to third parties if customer or patient data is compromised.

In short, AI‑enabled fraud collapses the traditional boundaries between HR, Legal, IT, Privacy and Security. Senior HR leaders can no longer treat hiring controls as an operational detail; they are part of the organization’s risk architecture.

The legal landscape: Federal pullback, state‑level acceleration

Complicating matters is a fragmented regulatory environment. At the federal level, enforcement related to disparate impact claims has narrowed, and prior guidance on AI in employment has been rolled back. Recent court decisions emphasize higher evidentiary thresholds for proving discrimination based on statistical disparities alone.

But for employers, this is not a green light—it is a trap.

States are rapidly filling the void with aggressive new laws governing AI in hiring and employment decisions. New York requires annual bias audits and public disclosures for automated employment decision tools. California has finalized sweeping regulations governing automated decision systems, including strict data retention requirements and prohibitions on discriminatory outcomes. Colorado now requires employers to exercise “reasonable care” to prevent algorithmic discrimination, backed by significant civil penalties.

For multi‑state employers, compliance is not about meeting a single federal standard. It requires designing AI governance to satisfy applicable laws in multiple states. To avoid liability, employers must satisfy the requirements of the most restrictive jurisdiction in which they operate, often before regulators or courts establish how these new laws should be interpreted and applied.

As states pass new, untested laws, the potential for liability is massive. Laws providing for set penalties for simple violations are a recipe for class action liability. Recent job posting disclosure laws have already led to hundreds of lawsuits, often resulting in six- or seven-figure payments to plaintiffs.

Why “black box” AI is a board‑level liability

One of the clearest messages for senior HR leaders is this: avoid black‑box hiring tools. If your organization cannot explain how an AI system evaluates candidates—what data it uses, how it was trained, and how decisions are made—you will struggle to defend it to regulators, courts, or your own board of directors.

Executives should expect AI vendors to provide documentation on training data, bias testing, auditability and regulatory compliance. Contracts with AI tool vendors must include audit rights, clear data‑use limitations and meaningful indemnification provisions. A vendor’s reluctance to provide transparency is itself a risk signal.

Importantly, HR leaders must also ensure that applicant data is protected. Privacy liabilities are expanding, particularly in California, where applicant and employee data receive full privacy protections. Mishandling resumes, video interviews, IP addresses, or inferred data can trigger fines, litigation and reputational damage.

Reinforcing the basics in an AI‑enabled world

Despite the advanced risks, many of the most effective defenses remain grounded in fundamentals. Camera‑on interviews, in‑person meetings where feasible, verified references, consistent background checks and zero‑tolerance policies for intentional misrepresentation matter now more than ever.

AI detection tools should be used cautiously. Poorly trained detection systems can introduce bias or trigger legal obligations of their own. Simple procedural safeguards—like requiring candidates to perform basic movements during video interviews—can be surprisingly effective without introducing new regulatory risk.

Most importantly, HR leaders must act quickly when fraud is suspected: suspend access, preserve data, involve IT and Legal immediately, and investigate thoroughly. Delay increases exposure.

The executive mandate

For senior HR executives, the path forward requires balance: leveraging AI’s legitimate benefits while building governance equal to its risks. This is not about slowing innovation—it is about ensuring that innovation does not outpace accountability and legal compliance.

The post The hidden risks of AI-driven hiring: How can HR leaders stay ahead? appeared first on HR Executive.

Date & Time: Thursday, June 25, 2026 at 2:00 pm ET

Speaker:

Jason Flynn, Human Capital Principal, Deloitte

Victor Reyes, Managing Director, Human Capital, Deloitte

Description:

How can senior HR executives help drive agility, value and growth within their mid-market organizations? Human capital management can help spur the growth these organizations need to thrive, but it’s important to understand the landscape and the trends — such as artificial intelligence — that are critical to success.

Join this free webcast, where Deloitte executives will share insights from the firm’s 2026 Global Human Capital Trends report and discuss the important choices HR leaders must make to optimize human performance in a rapidly changing world.

Attendees of this webcast will also receive access to exclusive research and insights from Deloitte’s recent research. Be sure not to miss this opportunity!

By attending, you’ll learn.

• Why three specific tipping points are redefining the future of work
• How human-machine collaboration can unlock stronger performance
• Which leadership choices separate reactive organizations from adaptive ones
• How dynamic orchestration is becoming essential in an AI-powered environment

Register here!

The post Global Human Capital Trends: 3 Ways to Super-Charge Mid-Market Growth appeared first on HR Executive.

Date & Time: Wednesday, June 24, 2026 at 11:00 am ET

Speaker:

Ariana Tiwari, Director of Product Marketing, Salesforce

Niamh McNamara, VP, Employee Services Global Operations, Salesforce

Zach Gray, Senior Director, HR Product Management, Salesforce

Description:

In an era where 72% of HR professionals see worker demands hitting record highs, Salesforce is closing the expectation gap by using its own agentic AI to reimagine HR. Join us to learn how Salesforce adopted Agentforce HR Service to achieve a 96% employee self-service resolution rate. We will go behind the scenes with our HR leaders to discuss how they unified fragmented systems like SAP, Workday and Slack into a single agentic platform that over the last 10 months has handled over 200,000 conversations and saved the HR team thousands of hours.

Key Takeaways:

• Learn how Salesforce achieved a 96% employee self-service resolution rate by using Agentforce HR Service to reimagine the employee experience.
• Discover how to unify fragmented systems like Workday, SAP, and Slack into a single agentic platform that executes tasks directly rather than just pointing to articles.
• Hear behind-the-scenes insights on how Salesforce saved the HR team thousands of hours and handled over 200,000 conversations in just 10 months.
• Explore the shift from reactive chatbots to proactive resolutions, including strategies for driving employee adoption by bringing AI into existing workflows like Slack.
• Understand the future of HR operations, including how to manage the employee-to-HR ratio and the necessity of upskilling humans to work alongside AI.

Register here!

The post How Salesforce Transformed Employee Experiences with Agentforce HR Service appeared first on HR Executive.

From data silos to institutional connectivity

The org employs about 150,000 people, operates in over 80 countries and manages 50+ brands across food, snacks and petcare. The company recently announced that it is deploying Google Cloud’s Gemini Enterprise as its primary AI operating system for employees worldwide, a rollout the company frames less as a technology upgrade and more as a fundamental change to how its workforce operates.

The recently announced partnership gives Mars employees (called Associates, internally) access to no-code and low-code tools to build their own AI agents tailored to their specific roles The goal, Mars says, is to turn decades of institutional knowledge trapped in data silos into something every employee can actually use.

For HR leaders watching how large enterprises are structuring AI adoption, the Mars model offers a concrete blueprint to an alternative to deploying AI as a top-down productivity mandate, as the company is positioning the move as a capability employees themselves can shape and direct.

“Our digital investments at Mars enable unlocking the potential of our Associates,” said Marina F. Bellini, global head of digital technologies at Mars and president at Mars Global Services, in a release.

Read more | Sabotage, silence and strategy ‘built for show’: 5 AI adoption myths

What HR leaders should pay attention to

The workforce design question embedded in Mars’ approach is one many CHROs haven’t fully answered yet: When employees can build their own AI tools, who governs what gets built?

Deloitte’s latest State of AI in the Enterprise report helps situate Mars on the broader adoption curve. Workforce access to sanctioned AI tools has jumped by about 50% in a year. This is up from under 40% of workers to roughly 60% now, as companies move from pilots to enterprise‑scale deployments.

Deloitte’s companion analysis on agentic AI finds adoption is racing ahead of oversight as only 21% of surveyed enterprises say they have mature governance in place for AI agents, even as usage is expected to rise sharply, with 74% of organizations forecasting at least moderate use of AI agents by 2027.

That gap reinforces why Mars’ emphasis on governance features is not just a technical detail but a differentiator in how it designs work. Mars says Gemini Enterprise’s governance features ensure every agent created across its petcare, snacking and food and nutrition segments is secure, compliant and consistent with company principles.

“This unified platform puts us in a great position to put business solutions at the core with technology as the enabler,” said Gülen Bengi, lead global chief marketing officer at Mars Incorporated and global chief growth officer at Mars Snacking. “We see incredible potential in our partnership as we bring solutions like One Demand AI to Mars—accelerating our growth pillars of innovation, brand building and end-to-end sales execution.”

The post Employee-driven AI at Mars: No-code agents unlock 115 years of knowledge appeared first on HR Executive.

A federal antitrust complaint filed April 2, 2026, alleges that BOLD Limited, through a web of related entities, controls more than 20 resume-builder brands. These include Monster, CareerBuilder, Resume Genius, My Perfect Resume, LiveCareer and Zety, and the complaint suggests that the entities use hidden dominance to run a deceptive subscription scheme on people actively looking for work.

Simulated competition

The case is Rocket Resume, Inc. v. BOLD Limited et al., No. 5:26-cv-02852, filed in the U.S. District Court for the Northern District of California. Plaintiff Rocket Resume, a small independent platform founded in 2019 and represented by Quinn Emanuel, alleges BOLD controls more than 80% of the U.S. online resume-building market, per the complaint, a market it values at more than $750 million annually.

According to the complaint, BOLD’s brands “offer functionally identical services, target identical customers, charge virtually identical prices, draw from identical content databases and operate under unified management control.” The cosmetic differences, such as color schemes, logos and domain names, are designed to simulate competition that does not exist, according to the accusation.

The complaint alleges that BOLD’s sites entice users with free or low-cost resume builders, then require a paid subscription to download the finished document. After the initial charge, users are billed 10 to 20 times that amount every four weeks, with cancellation made deliberately difficult, according to the filing.

Read more: Could video hiring solve problems that resumes never could?

’53 million Americans’

The complaint cites data showing 53 million Americans engaged in some form of job-seeking activity between November 2025 and January 2026. Much of this activity centers on resumes, as 97.8% of Fortune 500 companies filter applications through applicant tracking systems, per the complaint. A 2021 Harvard Business School study cited in the filing found that 88% of employers acknowledge their resume review process screens out qualified candidates at the initial stage.

“The result is that customers and job seekers are systematically ripped off by what appears to be a large range of online resume-building services, when in fact the choice is often Bold, Bold or Bold,” said Stephen Zimmerman, the software engineer who founded Rocket Resume, on a case information page. “This is unfair to job seekers and unfair to honest rivals like Rocket Resume.”

The complaint also alleges a longer pattern of competitive suppression. BOLD, it says, has filed copyright infringement suits against independent competitors for more than a decade, with some rivals ultimately absorbed into BOLD’s brand portfolio. Rocket Resume itself was sued by BOLD in 2022 and won partial summary judgment in May 2024.

Named individual defendants include Doug Jackson and Jamie Freundlich, BOLD’s co-chief executive officers, and Heather Williams, its former chief financial officer. The allegations have not yet been tested in court. BOLD had not issued a public response to the complaint at publication time.

The post Antitrust suit alleges Monster, CareerBuilder and Resume Genius are all the same company appeared first on HR Executive.

That’s how the best retained search professionals operate in an era of agentic AI, when the depth of relationships matters more than ever before.

For years, experienced recruiters have weathered one disruption after another—internal talent team dynamics, the onslaught of LinkedIn, allowing company HR teams to pop in keywords and treat talent like a search result and now AI. Each wave prompted the same question: Do recruiters still matter? And each time, the answer has been yes—but the reasoning has evolved. Today, the value of recruiters comes down to something simple but high-stakes: Organizations can’t afford to hire the wrong person.

See also: How AI platforms connect employers with executive job seekers

The resume isn’t the person

AI has greatly improved many aspects of recruiting. Algorithms can scan thousands of profiles, identify adjacent skills and surface qualified candidates at speeds no human can match. These tools help HR accelerate preparation and cull long lists of potential candidates and skillsets. But they can’t predict who’ll actually succeed.

Resumes disproportionately reflect opportunity, storytelling ability and historical context—not future suitability. Our research consistently shows that past performance is a poor predictor of success in larger or more complex roles without also assessing potential and learning agility. These capabilities—operating under pressure, leading through ambiguity, how a person fits within a company’s culture—are harder to see, even by the most skilled internal HR teams. That gap is where mismatched hires happen. I’m fond of saying, “You get hired for what you know and fired for who you are.”

Consider a controller who arrives at a financial firm with stellar credentials, excellent references and a history of transforming the function. Yet the culture, team dynamics and the pace of business at the firm don’t fit how she works. Within a year, she’s gone, and the organization is back at square one.

We see this all the time. Culture fit and potential are the two most important predictors of whether someone sticks around, earns promotions and grows with a business. They’re also the two things AI is least equipped to evaluate. A machine can identify who looks like the person who held the role before, but it can’t tell you who will excel in the role as it’s evolving.

This is where retained search thrives. Recruiters often build and maintain a trusted network cultivated through repeated interactions and candid conversations, allowing them to really understand who will watch your home and who may just say they’ll do it.

Merging science with instinct

The most effective retained search today uses data to validate what experience already suspects, and human judgement to interpret what the data can’t fully explain. Korn Ferry has analyzed more than 100 million assessments to understand what great looks like across four dimensions: competencies, experiences, traits and drivers. Of those dimensions, our data shows that agility captures the traits most relevant to a world in constant flux—tolerance for ambiguity, curiosity, adaptability and focus. Organizations that hire for agility dramatically outperform their competition. But that data, and the ability to gauge a candidate’s agility, still require interpretation by humans. That’s why we’re not surprised that as AI has become more embedded in search, we’ve seen retained search grow instead of shrink.

As agentic AI’s role continues to expand, the best outcomes won’t come from choosing between data and instinct. They’ll come from combining both. AI can accelerate some parts of search. But it takes a talent partner who has spent years building real relationships to know who will actually water the plants. That’s not a role the machines are coming for. It’s one they’re making more valuable.

The post Recruiters and AI agents: Co-pilots, not competitors appeared first on HR Executive.

Walmart is among several large employers, including Meta, Oracle and Amazon, that have announced workforce reductions in 2026.

Not all layoffs, as Walmart offers relocations

The cuts, according to WSJ, affect roles across tech, product and corporate support functions. Reports suggest employees are being offered relocation to Bentonville, Ark., or Northern California tech offices as Walmart centralizes distributed operations.

The reorganization was led by Suresh Kumar, Walmart’s head of global technology, who reviewed internal structures and identified overlapping roles created in part by the integration of AI-focused product units across Walmart, Sam’s Club and international divisions, according to The Wall Street Journal.

The company also recently brought on Daniel Danker, a former Instacart executive, to lead AI acceleration, a role that has been said to have directly shaped this restructuring. Walmart told WSJ the layoffs are not a direct result of AI replacing jobs, framing the move as organizational simplification.

As Donna Morris, Walmart’s chief people officer and HR Executive’s 2025 HR Executive of the Year, has put it, Walmart’s AI work is meant to build a “people-led and tech-powered” organization, a deliberate choice she says has informed every AI decision HR has made.

Read now: As companies use AI to justify layoffs, Josh Bersin says HR is solving the wrong problem

Walmart’s AI strategy

Walmart has spent the past several years building out one of the more closely watched AI workforce strategies among large employers. In August 2025, HR Executive reported that the company had consolidated dozens of AI agents into four “super agents” covering customers, employees, engineers and sellers to simplify the associate experience across its 2.1-million-person global workforce.

Former CEO Doug McMillon, who retired in January after four decades with the company, earlier this year said that the pace of AI change influenced his decision to step down. “With what’s happening with AI, I could start this next big set of transformations with AI, but I couldn’t finish,” McMillon said, as reported by HR Executive in March.

The relocation component of Walmart’s announcement deserves particular attention. Requiring employees to move to a centralized hub rather than offering severance or remote options is a workforce strategy choice that carries retention risk. This may signal a move away from distributed work for corporate tech roles and raises the stakes of change management. Reuters noted that relocation is central to how Walmart is executing this reduction.

Walmart’s fiscal Q1 2027 earnings are scheduled for May 21 where further details about workforce strategy and AI investment are expected.

The post Walmart cuts 1,000 corporate roles as AI reorganization reshapes tech teams appeared first on HR Executive.

The annual report, which surveyed about 300 C-suite executives, found that 84% of respondents expect AI-related policy or regulatory changes in the next year. That figure is double what it was at this time last year.

Niloy Ray, co-chair of the AI & Technology Practice Group at Littler, says that the shift highlights a number of realities, including “natural growth” in the number of businesses that have implemented AI over the last year.

“This year, we’re closer to AI becoming the median expectation, as most employers are using it and grappling with regulatory and compliance obligations,” Ray says.

At the same time, the regulatory landscape is growing more complex. Multiple measures are moving forward to regulate AI in a number of states, while the Trump administration continues to press for a uniform approach that limits state oversight.

“Smart employers are expecting more and more change,” Ray says of the regulatory landscape.

Room for improvement on AI governance

Despite the pace of AI adoption, governance is lagging.

The report finds that 68% of organizations do have a formal policy around AI use in the workplace, nearly double where the figure stood last year. However, such policies are often not comprehensive enough, Ray says. Just over half have restrictions on the information that can be entered into the tools, while approximately the same percentage have a formal process to approve the tools and use cases.

Without such provisions, organizations could be damaging both the AI confidence and the utilization of the workforce.

“Too many businesses, in my experience, don’t provide a clear enough pathway to using AI, because their governance and policies tend to put the compliance and risk mitigation burden on the end user, rather than the organization as a whole shouldering that,” Ray says.

For instance, when policies dictate to employees that their AI use cannot subject the organization to bias, hallucinations or plagiarism, it’s unrealistic to expect end users to figure out such risks on their own. Employees should be empowered to use AI, Ray says, in a way that matches the organization’s “enthusiasm for AI use and level of risk aversion.”

Policies should be created through a “multi-stakeholder process,” involving a governing group from leaders across functions. Guidelines should stipulate approval, implementation, use and development of AI and must include very clear, “not prospective, but descriptive” objectives.

At the end-user level, it’s vital to have “clear do’s and don’ts,” Ray says. A handbook, for instance, should cover which AI tools are approved, processes for seeking approval outside of that list and basic reminders on how to treat confidential information.

“All of those basic rules of the road are what the company as a whole has decided on—and then it’s made clear to employees that this is what the philosophy is of the business, which has taken on the burden of deciding which tools and uses meet expectations,” Ray says. “Then, employees can go forward empowered, knowing the business has taken on the burden of managing risk.”

See also: What an employment attorney wants HR leaders to know about AI risk

A rising role for HR

When it comes to implementation, HR appears to be ground zero for most companies, reflecting the “critical” role CHROs play in shaping the development and rollout of AI strategy.

When asked which areas of the organization AI was being used in, HR topped the list—tied with IT—at 54%. Following these two functions, AI use is most common in marketing, sales and communications; followed by legal and compliance.

HR is the lynchpin that takes AI policy from being hypothetical to a reality, Ray says.

“When the rules meet reality, if they haven’t been tested against what employee workflows and expectations are, people won’t use it,” he notes. “CHROs understand how AI is going to be used.”

They also represent the function that “knows how to beat to the collective heartbeat of its people,” which will be key to driving adoption.

“We can’t forget,” Ray says, “the human part of the equation is the hardest one to manage and also the most critical to protect.”

The post Littler: AI overtakes immigration, DEI as top employer concern appeared first on HR Executive.