Robert Nickelsberg/Getty Images
- Leaders and employees don't always agree on who's responsible for AI upskilling.
- Because the technology is changing so fast, traditional corporate training won't cut it.
- "We all have to learn a new thing, even if it means doing it on our own time," one CEO said.
Larry Gadea wants the people who work for him to stay up to speed on AI. That might mean doing some homework.
The founder and CEO of the workplace software company Envoy said the company is investing in AI training to help employees embrace the technology as a tool that can help them do better work. Yet he also believes workers have to take responsibility for learning outside the office.
"We all have to learn a new thing, even if it means doing it on our own time," Gadea said.
As AI rewrites the instruction manual for job categories such as software development, a new question is emerging: Who is responsible for making sure workers keep up?
While many business leaders now view AI fluency as a baseline skill, bosses and workers are sometimes divided over how much responsibility belongs to employers and how much belongs to employees themselves.
That tension showed up in a recent survey by the management and technology consultancy Emergn. About eight in 10 CEOs said employees should take responsibility for upskilling themselves, while a similar share of employee respondents said companies should provide AI training.
At Envoy, which helps companies manage office visitors and desk and room reservations, employees often demonstrate AI tools they've built and explain how they developed them during all-hands meetings, generally twice a month. Individual teams also meet regularly to discuss ways AI can improve workflows, such as automatically preparing account histories for sales calls.
The point, Gadea said, isn't to prove expertise. It's to normalize experimentation. Gadea said he frequently uses the term "imposter syndrome" to remind employees that the tech is new for everyone.
"You're literally saying, 'Hey guys, it's OK. We're all learning together,'" he said.
The challenge of formal training
Figuring out how to teach workers about AI has been one of the biggest management challenges in recent years. Companies have tried a variety of methods during the traditional 9-5, including hackathons, training weeks, and dashboards to illustrate AI use. Some workers are spending nights and weekends running experiments they can show off when they're back at work.
Of course, not all workers are as excited about AI as the boss might be. Research shows a gap between employees' and employers' eagerness to embrace tech.
In practice, workplace learning has long been a shared responsibility between employers and employees. About 60% of workplace learning tends to be self-directed through resources like articles and videos, said RJ Bannister, the chief operating officer at the executive compensation and corporate governance consultancy Farient Advisors.
Another 30% comes from hands-on experience, including working with colleagues and on projects, and about is 10% from formal training sessions, he said.
AI can make the process more efficient by letting people work through more topics in less time, Bannister said. Yet, he added, workers still need to apply what they learned to make it stick.
Bannister compares worker training and development to how a company spends money to maintain factory equipment.
It's another way, he said, of making workers feel that when it comes to new technology like AI, "they're a part of a solution, as opposed to just seeing it solely as a threat."
At the private crypto exchange GoDark, founder and CEO Denis Dariotis said many employees simply learn by doing. In part, that's because while they might be experienced software engineers, most didn't start their careers in crypto trading, he said.
"There is a lot of our team that's sort of just thrown into the deep end, and then they have to figure it out," Dariotis said.
GoDark doesn't require formal AI training for its roughly three dozen workers. To keep up, employees share articles and ideas in Slack, sometimes after hours, though Dariotis, who also runs the crypto infrastructure firm GoQuant, said he doesn't expect people to spend their evenings learning after regularly putting in 12-hour days. At the same time, Dariotis said he doesn't mind seeing it.
"I think everyone has this natural interest in the work," he said.
Learning at the pace of AI
Aside from limited time in the workday, another challenge is ensuring that any company training stays current.
Companies can't rely on quarterly training sessions when new AI models and capabilities arrive every few weeks, said Kathy Gersch, CEO of the change-management firm Kotter.
"You're not training on a process or how to use a technology," she said, adding that what is current this month might be different by the next.
Instead of relying solely on formal classes, companies should build systems that help employees continuously share what they're discovering with one another, Gersch said.
When corporate training is offered, it's often too generic and not tied to a company or a job, said Mark Ma, an associate professor at the University of Pittsburgh School of Business. That can make it less useful.
For Gadea, learning AI isn't only about succeeding at Envoy. Workers who want to move to another company, he said, will increasingly be expected to demonstrate proficiency with AI tools.
Gadea said that's yet another reason to put in the work — both on and off the clock.
Do you have a story to share about your AI experience at work? Contact this reporter at tparadis@businessinsider.com.
Read the original article on Business Insider