AI can't improve company culture. But it empowers employees to try

By now, organizations have a much better understanding of the types of tasks and roles they can offload to artificial intelligence. But chatbots won't be replacing more human-centric roles anytime soon. 

According to a working paper from the Massachusetts Institute of Technology, the use of hyper-intelligent chatbots is raising productivity, decreasing inequality between workers and enhancing job satisfaction. In fact, 75% of respondents who implemented ChatGPT saw improvements in team morale, collaboration and collective learning. While there are positives to this partnership between AI and employees, there will always be a limit. 

"Everything's going to have an effect on culture," says Danny Gutknecht, CEO of Pathways, a human resources platform. "AI is definitely going to change a lot about jobs as we know them, but it's not going to replace core human creativity, communication and connection — all of the things that make a culture." 

Read more: 49% of CEOs say AI could do the bulk of their work. What does it mean for the C-suite?

Companies across every industry have been experimenting with where AI could improve their business. According to a 2022 study by IBM, 26% of organizations are using AI for marketing and sales, while 52% are using it to improve IT, with overwhelmingly positive results. In fact, 48% of organizations are seeing a better customer experience, according to IBM, quelling much of the initial trepidation and fear employees and employers alike had about rolling out more tech tools. 

"AI is helping improve cognitive functions and can do a better job of thinking logically about things," Gutknecht says. "It might even free us up to be more effective at doing the things that humans do really well." 

But in order to maximize this effect, employers will first need to understand the areas they shouldn't be automating, as much as they've considered those they should, according to Gutknecht. For example, recruiting and data analysis are good places to start for teams looking to expand their AI use. But when it comes to the creation of new policy or engagement strategies, it's best to keep the methods more traditional. 

Read more: A step-by-step guide to implementing AI at work

"Humans get better at things and evolve by engaging in conversations," Gutknecht says. "Those conversations have some sort of tension or conflict, or some sort of dynamic because that is how things get innovated. And that's one of the things that I don't see AI being able to ever do."

Moving forward, organizations should be setting boundaries as they implement more AI. Gutknecht suggests that employers prioritize employees' ability to communicate with each other and with management. He urges companies to follow-up with employees once they've automated certain administrative tasks in order to understand how that automation may have created room for more connection and engagement. 

"Whether you're a leader or whether you're just somebody working within a company's culture, there's a connectedness," he says. "If we outsource all of that to AI, what happens to that capacity? Would we become much more shallow and lose depth? What if we instead used AI as just an assistant to gain depth? We should be looking for the answer to that."