Amazon employees told use AI or else... they started faking tokenmaxxing

A Financial Times report reveals Amazon employees are allegedly faking AI tool usage, such as 'MeshClaw,' to inflate metrics and meet internal adoption targets, despite the company denying direct performance review ties. Amazon plans to spend $200 billion this year on AI and data centers, pushing tools like MeshClaw—an AI agent platform—to automate tasks like code deployment and email management, mirroring the design of OpenClaw." "article": "Amazon employees are reportedly manipulating internal AI usage metrics to comply with the company’s push for broader adoption. According to a Financial Times report, some workers are using the company’s internal AI tool, MeshClaw, for unnecessary tasks to inflate their 'token consumption'—a measure of AI processing. This practice suggests employees fear informal judgment from managers despite Amazon’s official stance that AI usage won’t directly impact performance reviews. MeshClaw, an AI agent platform inspired by OpenClaw, automates workplace tasks such as deploying code, managing emails, and interacting with tools like Slack. Amazon has expanded its use across the company, aiming for over 80% of developers to engage with AI tools weekly. The company tracks AI usage through internal leaderboards, though it insists token statistics won’t factor into evaluations. The report highlights Amazon’s aggressive AI investment, with an estimated $200 billion allocated this year for AI technology and data center infrastructure. Employees believe the pressure to adopt AI tools stems from Amazon’s broader strategy to integrate AI into operations, even if it leads to artificial metric inflation. Critics argue the focus on AI adoption metrics creates perverse incentives, pushing workers to game the system rather than use AI meaningfully. Meanwhile, Amazon’s push for AI-driven productivity continues, with tools like MeshClaw central to its vision for a fully automated workplace.