AI Compute Costs Drive Shift To Usage-Based Software Pricing
Displaying 1-1 of 1
| Yesterday, 12:36 | |
duuhdeek@gmail.comTotal Topics: 0 Total Posts: 0 |
The software-as-a-service industry is undergoing a fundamental transformation, abandoning the decades-old "per seat" licensing model in favor of usage-based pricing structures. This shift, Business Insider reports, is primarily driven by the astronomical compute costs associated with new "reasoning" AI models that power modern enterprise software. Unlike traditional generative AI, these reasoning models execute multiple computational loops to check their work -- a process called inference-time compute -- dramatically increasing token usage and operational expenses. OpenAI's o3-high model reportedly consumes 1,000 times more tokens than its predecessor, with a single benchmark response costing approximately $3,500, according to Barclays. Companies including Bolt.new, Vercel, and Monday.com have already implemented usage-based or hybrid pricing models that tie costs directly to AI resource consumption. ServiceNow maintains primarily seat-based pricing but has added usage meters for extreme cases. "When it goes beyond what we can credibly afford, we have to have some kind of meter," ServiceNow CEO Bill McDermott said, while emphasizing that customers "still want seat-based predictability." EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA EhhA |
Displaying 1-1 of 1
