metamorworks - stock.adobe.com

Conflicts, questions over generative AI at SHRM

At the SHRM conference this week, the increased use of AI, especially generative AI, in HR was a major focus. Discussions centered on skills, regulation and use cases.

The escalating adoption of AI by HR departments, especially generative AI, was a key topic at the annual Society for Human Resource Management conference in Las Vegas this week. The discussion examined the transformative shift in the requisite skills for an AI-reliant future.

Just seven months into its deployment, generative AI tool ChatGPT has made inroads into HR technologies and operations. But forecasting the implications of rapidly evolving AI tools remains a challenge, a persistent theme throughout the conference.

HR professionals are using generative AI for tasks such as creating employee handbooks and writing job descriptions, said Kim Crowder, an independent HR consultant who spoke at the conference on AI's use in HR. But there are concerns about security and copyright infringements when using AI-generated content. She said compliance is also an issue because ChatGPT could generate content for an employee handbook that's not in alignment with state regulations.

"The truth of the matter is we're all learning," Crowder said in an interview, "and it's moving quickly." 

Among the top concerns expressed in multiple conference sessions is the need for AI-related skills to remain relevant in one's job. Another issue is the fast-moving AI regulations from states and local governments, which may already discourage using some tools. 

New AI audit regulations in New York City, for example, may prompt nonprofit community healthcare facilities to end their use of AI, said Ken Meyer, a New York-based HR consultant who works with healthcare companies. New York City's new independent automated employment decision tool law, called NYC Local Law 144, requires companies to conduct annual audits of decision tools they use for hiring. 

Regulation is hurting use

"We've pretty much been regulated out of using these tools," Meyer said during a conference session, "because the money we would have to invest in the auditing, that's money that was taken away from our patients."

Meanwhile, the broader legal landscape for AI is shifting rapidly. Anjelica Dortch, senior director of U.S. government affairs at SAP, said that more than 150 AI-related bills had been introduced in states this year alone. 

In the global context, Dortch said that European lawmakers would likely finalize AI regulations before their U.S. counterparts, possibly within the year. U.S. lawmakers are moving more slowly, an approach that Dortch endorses because it lowers the risk of hasty and potentially restrictive rules. 

"I do not expect Congress to pass comprehensive AI legislation this year -- they're just not ready," she said at a conference session.

A frequent sentiment expressed throughout the Society for Human Resource Management (SHRM) conference was the necessity for HR pros to educate themselves on AI functionality. 

AI isn't going to replace your job, but the person who will replace you will have your same expertise and will also understand how AI works.
Megan Smith-Branch Deputy lead of the AI ethics and safety team, Booz Allen Hamilton

Megan Smith-Branch, deputy lead of the AI ethics and safety team at Booz Allen Hamilton, said in a presentation that HR professionals should understand how AI is relevant to their profession.

"AI isn't going to replace your job," Smith-Branch said. "But the person who will replace you will have your same expertise and will also understand how AI works."

Generative AI skills for HR that emerged during the presentations weren't necessarily focused on how to program algorithms as much as on how to apply generative AI to tasks and understand data, the risks -- such as bias -- and the benefits of the tool.

In a separate interview, David Wagner, senior research director at Avasant's Computer Economics research group, concurred and predicted that AI knowledge would become a standard skill in every job, like using email.

Wagner said he expects to see the emergence of "AI-native" workers in the next five to 10 years.

"We'll be talking about kids who grew up doing their homework side-by-side with digital assistants and generative AI," Wagner said. "When they enter the workforce, that's when we'll see real change."

Patrick Thibodeau covers HCM and ERP technologies for TechTarget Editorial. He's worked for more than two decades as an enterprise IT reporter.

Dig Deeper on Talent management