Venture capitalist and LinkedIn co-founder Reid Hoffman has publicly endorsed the practice of tracking employee AI usage, known as "tokenmaxxing," as a critical step for companies adapting to the artificial intelligence age. His comments, made during an interview at Semafor's World Economy summit, come days after internal controversy at Meta over similar tracking dashboards.
An AI token is a unit of data processed by an AI model to understand prompts and generate responses. It is also the primary metric for measuring usage and cost on AI platforms. The concept of "tokenmaxxing" – where "maxxing" is Gen Z slang for optimising something – involves companies internally ranking employees based on their token consumption as a proxy for AI engagement.
Support Amidst Silicon Valley Debate
Hoffman expressed a favourable view of the practice, stating it provides a valuable, if imperfect, dashboard for understanding how an organisation is experimenting with AI tools. "You should be getting people at all different kinds of functions actually engaging and experimenting [with AI]," Hoffman said. "Here’s one of the things that is a good dashboard to be looking at — doesn’t mean it’s a perfect example of productivity, but… how much token usage are people actually doing as they’re doing it?"
He cautioned that high token usage alone is not a definitive measure of productivity, noting that some usage may be exploratory. The key, according to Hoffman, is to pair the metric with an understanding of what employees are using the AI to accomplish. "Some of it will be experiments that’ll fail — that’s fine. But it’s in that loop, and you want a wide variety of people using it essentially, collectively, and simultaneously," he added.
Broader AI Strategy Advice
Beyond tracking usage, Hoffman advised that AI should be embedded across entire organisations. He recommended implementing regular check-ins, such as weekly meetings, for teams to share discoveries and successful applications of AI. "What did we try to do new this week, to use AI for both personal and group and company productivity, and what did we learn?" he suggested as a framework, adding that such practices would uncover "really amazing" efficiencies.
The endorsement arrives amid a heated debate within the tech industry. Critics argue that ranking employees by token spend is a flawed metric, akin to measuring productivity by who spends the most money. The practice gained widespread attention after reports that Meta had shut down an internal "tokenmaxxing" leaderboard following a press leak.
Context and Industry Reaction
The controversy highlights the growing pains of integrating generative AI into corporate workflows. As companies invest heavily in AI subscriptions, many are seeking tangible metrics to gauge return on investment and employee adoption rates. Hoffman's stance lends significant credibility to the "tokenmaxxing" approach, positioning it not as a punitive measure but as a diagnostic tool for fostering a culture of experimentation.
Industry observers note that the discussion reflects a larger strategic shift. Some analysts suggest that internal tracking of AI usage, as seen at Meta, may indicate a move towards greater vertical integration and development of proprietary AI models to reduce reliance on external, token-based services from companies like OpenAI and Anthropic.