Reid Hoffman Backs Controversial "Tokenmaxxing" AI Employee Tracking Trend

Just days after Meta shuttered its internal "tokenmaxxing" dashboard following a leak of the company’s private AI leaderboard to the press, LinkedIn co-founder and prominent venture capitalist Reid Hoffman has publicly thrown his support behind the viral concept that has recently swept across Silicon Valley tech circles.

To set context, an AI token is a small, discrete segment of data that large language models process when interpreting user prompts and generating output. Tokens also serve as the standard industry unit for measuring AI platform usage, and form the basis for how most AI service providers calculate customer costs for their services.

As organizations rush to roll out generative AI tools across all internal teams, many companies have begun tracking which employees consume the most tokens, using that metric as a proxy to identify workers who are most actively embracing and experimenting with new AI tools. The practice has been dubbed "tokenmaxxing" — the "maxxing" suffix draws from Gen Z slang for intentional optimization, a turn of phrase already used in popular terms like "looksmaxxing" and "sleepmaxxing."

The trend has sparked heated debate among tech engineers, who question whether token usage counts as a viable, fair measure of workplace productivity. Critics argue that tracking token consumption is little different than ranking employees based on how much company budget they spend, regardless of the output or results that spending generates.

Speaking in an interview at Semafor’s World Economy Summit this week, Hoffman shared guidance for companies integrating AI into daily operations, and made clear he holds a favorable view of the token tracking practice. Though he did not reference the metric by its Gen Z slang name, he explicitly stated that monitoring employee token spend is a smart strategy for organizations.

“You should be getting people at all different kinds of functions actually engaging and experimenting [with AI],” Hoffman said at the event. “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?”

Hoffman went on to note that high token usage does not always correlate to strong work output: some employees may use large volumes of tokens for unstructured, open-ended exploration, which is why organizations should pair tokenmaxxing tracking with clear context about what projects employees are using their tokens for.

“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,” Hoffman added.

Hoffman also shared additional advice for companies navigating the process of building out organizational AI strategies. He recommended that AI tools be embedded across every department of a business, rather than confined only to tech or product teams, and that companies hold regular check-ins for employees to share successful tactics and key learnings with one another.

“We should have, essentially, a weekly check-in. It doesn’t have to be everyone, all the time with each other –but a group check-in about ‘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?’ Because what you’ll find, some of the things are really amazing,” Hoffman said.