The AI Training Dilemma: Who is Responsible for Upskilling Workers?
As artificial intelligence rapidly transforms the modern workplace, a new tension has emerged between employers and employees: Who is actually responsible for keeping workers up to speed on AI?
For Larry Gadea, founder and CEO of workplace software company Envoy, the answer requires a shared effort. While his company invests in AI training to help staff embrace the technology, Gadea firmly believes that workers must also take the initiative to learn on their own time. "We all have to learn a new thing, even if it means doing it on our own time," he stated.
This perspective highlights a growing divide in the corporate world. As AI rewrites the rules for roles ranging from software development to marketing, business leaders increasingly view AI fluency as a baseline requirement. However, a recent survey by the consultancy Emergn reveals a stark disconnect: roughly 80% of CEOs believe employees should be responsible for their own upskilling, while an equally large share of employees argue that companies should provide the training.
Normalizing Experimentation
At Envoy, the focus is on making AI a regular part of the conversation. During bi-monthly all-hands meetings, employees frequently showcase AI tools they’ve built and explain their development process. Individual teams also hold regular discussions on how AI can streamline workflows, such as auto-generating account histories for sales calls.
According to Gadea, the goal isn't to showcase expertise, but to normalize experimentation. By openly acknowledging that everyone is navigating uncharted territory, leaders can help alleviate the "imposter syndrome" many workers feel. "You're literally saying, 'Hey guys, it's OK. We're all learning together,'" Gadea explained.
The Reality of Workplace Learning.
While some companies host hackathons, training weeks, or create internal dashboards to track AI usage, the reality is that professional development has always been a collaborative effort. RJ Bannister, COO of Farient Advisors, notes that workplace learning typically breaks down into three categories:
- 60% self-directed: Employees learning through articles, videos, and independent research.
- 30% hands-on: Learning through experience, collaborating with peers, and working on projects.
- 10% formal: Structured training sessions.
Bannister compares employee development to maintaining factory equipment—it's an investment that makes workers feel like active participants in a solution rather than viewing new technology as a threat.
Learning in the "Deep End"
Not every company has a formal AI curriculum, and some employees are more enthusiastic about the tech than others. At GoDark, a private crypto exchange, CEO Denis Dariotis says his team largely learns by doing. Because many of his experienced software engineers are new to the crypto trading space, they are often "thrown into the deep end" to figure things out.
GoDark doesn't mandate formal AI training for its roughly three dozen employees. Instead, the team shares articles and ideas in Slack channels, sometimes after hours. While Dariotis doesn't explicitly expect his team to study after grueling 12-hour shifts, he notes that their natural interest in the work often drives them to do so anyway.
The Problem with Formal Training
Relying solely on structured corporate training may be a losing battle when it comes to AI. Kathy Gersch, CEO of the change-management firm Kotter, points out that new AI models and capabilities emerge every few weeks, making quarterly training sessions obsolete almost as soon as they are delivered.
"You're not training on a process or how to use a technology," Gersch explained, emphasizing that companies need to build systems that encourage continuous peer-to-peer knowledge sharing instead.
Furthermore, when formal training is provided, it often misses the mark. Mark Ma, an associate professor at the University of Pittsburgh School of Business, notes that corporate AI training is frequently too generic and disconnected from specific company needs or individual job roles, severely limiting its usefulness.
The Long-Term Payoff.
Ultimately, the push for AI proficiency isn't just about benefiting the current employer. Gadea points out that as the job market evolves, workers who want to advance or move to a new company will increasingly be expected to prove their competence with AI tools.
In a rapidly shifting technological landscape, putting in the work—whether on the clock or off it—may be the only way to future-proof a career.