AI adoption may be entering a more cost-efficient phase. According to TechCrunch, AI spend per employee at top firms slipped in August, with falling token costs and cheaper models playing a key role.
For businesses, that decline does not have to be bad news. Lower AI costs can mean teams are able to access useful capabilities with less budget pressure, potentially making experimentation and deployment easier to justify.
A shift from spending to efficiency
The trend may be less exciting for hyperscalers hoping for ever-rising AI infrastructure revenue, but it points to a healthy dynamic for customers: competition and technical improvements are making AI cheaper to use.
- Cheaper models can reduce barriers to adoption.
- Lower token prices can make everyday AI workflows more affordable.
- Businesses may begin judging AI more by productivity gains than by how much they spend.
If the pattern continues, the real win could be a more practical AI market—one where organizations focus on sustainable value rather than headline-grabbing budgets.