Amazon just scrapped an internal leaderboard that tracked how much employees used AI tools. The company’s senior executive Dave Treadwell told staff to stop using AI just for the sake of using it.
The move comes as AI costs are rising and Amazon realized workers were gaming the system. Instead of using AI to actually help their work, employees were chasing high usage scores to climb the leaderboard rankings.
The AI Usage Problem
Amazon created the leaderboard to encourage workers to try new AI tools and boost productivity. But it backfired spectacularly. Employees started using AI for everything – even tasks where it didn’t make sense – just to rack up points.
This created two problems. First, it drove up Amazon’s AI bills as workers used the technology unnecessarily. Second, it didn’t actually improve work quality or speed. People were using AI to write emails that took longer to review than writing by hand, or asking it questions they already knew the answers to.
Treadwell’s message was clear: use AI when it genuinely helps, not because there’s a scoreboard watching.
What This Means
Amazon’s experience shows how tricky it is to roll out AI in big companies. The technology is powerful, but getting people to use it smartly takes more than just tracking numbers.
Other companies watching Amazon’s AI experiment will likely learn from this mistake. Expect to see fewer usage-based incentives and more focus on actual results when businesses deploy AI tools to their workers.
