With the rapid integration of AI technologies in workplaces, managers worldwide are grappling with the challenge of assessing employee proficiency in using these tools. Companies are exploring various methods to incorporate AI usage into performance reviews, though the path forward remains uncertain.
AI Archetypes: A Framework at Gusto
At Gusto, a human resources platform, managers evaluate employees’ AI usage by categorizing them into five archetypes during quarterly reviews. Workers may start as “observers,” not yet utilizing AI independently, and progress to “integrators,” who incorporate AI into their daily tasks, achieving consistently better outcomes. The most advanced category, “amplifiers,” involves employees who innovate new AI applications and share insights with colleagues.
Scott Helmes, Gusto’s chief people officer, emphasizes the initiative’s aim to standardize AI implementation across the company rather than enforce strict rules. “We want to be able to have a coaching conversation,” he explains. The next objective is to more directly assess the impact of AI on workers’ contributions. “The introduction of this, I think, has really helped us have a consistent way to talk about AI fluency across the organization,” Helmes adds.
Shifting Strategies and Industry Reactions
The grading rubric at Gusto emerged amid a broader industry trend known as tokenmaxxing, but faced swift criticism. Companies like Amazon and Uber initially employed leaderboards to monitor AI use, but have since shifted away from maximizing AI usage. Similarly, Duolingo retracted plans to include AI use in performance evaluations. Meanwhile, Meta faces legal challenges over allegations of using AI to rank employee productivity, which the company denies.
The Complexity of AI in Performance Reviews
The quest to separate AI superusers from those who avoid it has become a part of performance reviews. However, the lack of clear expectations means employees strive to meet undefined standards. Richard Landers, a University of Minnesota professor, notes the unfair pressure on frontline workers. “There’s a lot of drive to push that responsibility onto frontline workers in a way that’s not particularly fair,” he states.
AI’s integration into performance reviews is often subjective, as illustrated by a survey from General Assembly, which found that nearly half of 500 business leaders in the US and UK had incorporated AI into employee assessments. The survey focused on AI tool usage, performance improvements, and anecdotes of enhanced efficiency.
Industry Perspectives and Challenges
Without concrete benchmarks like tokens spent or lines of code generated, companies often assess AI use by tracking increased productivity. Stefan Camilleri, Typeform’s vice president of engineering, queries engineers: “Now that I’ve given you a new tool, how much faster are you doing it?” Yet he expects more than just speed. Shensi Ding, CEO of Merge, evaluates employees based on the quality of their AI use, rewarding those who effectively teach others.
Startups and large corporations alike are scrutinizing AI use. Meta included “AI driven impact” in its performance reviews, while Accenture tracks AI logins for promotions. Even at Google, AI proficiency may appear in evaluations, though it’s not mandatory.
Lee Senderov, Travelport’s chief transformation officer, highlights the importance of organic AI adoption. Her company uses AI to analyze code, providing insights for discussions about workers’ understanding and coding practices. “We want this to happen organically and happen through inspiration,” she explains.
Redefining Performance Metrics for AI
The lack of clarity around AI’s role in performance reviews has led to inconsistent evaluations. A Harvard Business Review article suggests focusing on workers’ ability to evaluate AI accuracy, improve team productivity, and adapt to new technologies and workflows.
Despite this, companies have not widely standardized these changes. A Deloitte report found that 84% of firms have yet to redesign work to align with AI capabilities. “Leaders are realizing that this usage conversation isn’t having the impact that they were looking for,” says Lori Moffatt from Waterstone’s Leadership and Culture Advisory Services.
Marc Cenedella, founder of Ladders, urges patience as AI evolves rapidly. He encourages experimentation with AI, even if current efforts seem trivial. “It kind of doesn’t matter if it’s useless or useful at this point,” Cenedella remarks. The transformative potential of AI remains a work in progress, awaiting streamlined integration across industries.
