‘Just keep prompting’ is costing big tech millions — the rise of AI ‘tokenmaxxing’

What Is ‘Tokenmaxxing’?

The growing concern is linked to a term increasingly used in AI circles — “tokenmaxxing”.

In the AI ecosystem, tokenmaxxing refers to maximising prompts, outputs, and workflows to extract greater value from AI models.

However, excessive prompting and overuse of AI tools by employees have reportedly led to soaring AI-related expenses for several technology firms.

As companies expand the use of generative AI across departments, uncontrolled usage is significantly increasing computing and API costs.

Companies Begin Tightening AI Budgets

The rising expenditure has prompted major firms to introduce stricter AI budgeting and internal controls to prevent costs from spiralling out of control.

According to reports, Microsoft has cancelled its Claude Code licences for employees and asked them to shift to the company’s internal AI tools by June 30.

Similarly, Uber COO has reportedly warned that the “just keep prompting” culture is no longer a sustainable business model due to escalating AI costs, Business Today reported.

Amazon Flags ‘Tokenmaxxing’ Behaviour

Amazon has also reportedly scrapped an internal leaderboard that tracked employee AI usage after discovering that some workers were inflating their numbers with unnecessary AI activities, increasing the company’s computing costs, Financial Times reported.

Dave Treadwell, Amazon’s senior vice-president, reportedly told staff that while the leaderboard was created with good intentions, employees’ “tokenmaxxing” behaviour ultimately drove up expenses for the company.