Token leaderboards meet pushback at Web Summit Rio

token leaderboards – At Web Summit Rio, Replit’s Michele Catasta denounced the corporate practice of ranking employees by how many AI tokens they use as “very dystopian,” arguing token consumption is not tied to real impact and can drive waste. The backlash follows moves by Amazon

For a while, “AI usage” inside offices sounded like a clean metric: more tokens consumed, more adoption proven. Then Michele Catasta started calling out the logic that turned workplaces into leaderboards.

Speaking at Web Summit Rio this week. the president and head of AI at Replit described a trend where companies build internal dashboards that make people compete to climb rankings based on how many AI tokens they use at work. In his view, the setup turns productivity into consumption. “People are basically competing to be at the top of the ranking and showing everyone that they’re burning more tokens than anyone else. ” he said.

Catasta didn’t soften the message. “That is very dystopian,” he said.

The pushback is landing as more executives start challenging whether maximizing token usage is actually measuring what matters—impact. not activity. Catasta argued token consumption is a poor gauge of performance inside a company. “Token usage is not proportional to the amount of impact that those people are having within the company,” he said.

He also pointed to the incentives such systems can create. Even if the original goal is tracking adoption, he warned that the approach can encourage wasteful behavior. “It’s also irresponsible. ” he said. comparing excessive AI use to leaving the lights on at home and not caring about the electricity bill. He added that using AI excessively carries real costs: “more energy is being used. ” and there is “less capacity for other use cases that other companies want to build on top of.”.

Catasta’s comments arrive right as companies publicly distance themselves from tokenmaxxing—the term critics use for chasing higher AI spend or higher token totals for its own sake.

Last month, Amazon scrapped its internal AI-use leaderboard. A spokesperson said it “was never intended to promote the use of AI for usage’s sake.” Around the same time, Uber COO Andrew Macdonald said he wasn’t seeing proportional productivity gains from increasing AI costs.

BNP Paribas CIB added its own warning signal through its chief AI officer, Charles Holive, who dismissed tokenmaxxing as a “vanity metric.”

The thread running through the criticism is consistent: dashboards that reward higher consumption can distort the way teams allocate budget and effort. Catasta framed it as an accountability failure—measuring “burning more tokens” rather than outcomes—while also tying the argument back to operational limits. including energy use and reduced capacity for other work.

For now, the story is less about whether companies use AI and more about how they judge success. As more executives pull away from token leaderboards and “usage for usage’s sake,” the question is shifting from adoption volume to business value—and whether internal metrics are capable of catching up.

Replit Michele Catasta Web Summit Rio token leaderboards AI ROI tokenmaxxing Amazon Uber BNP Paribas CIB Andrew Macdonald Charles Holive energy use AI adoption metrics