AI companies rethink pricing as high running costs hit budgets

In short: AI tools are costing a lot to run, so AI companies are rethinking how they price their products and services.

What's going on

Many of today’s strongest AI systems are expensive to operate. Each time you ask an AI a question, it uses computing power in a data center, which costs money for chips and electricity. Think of it like using a taxi meter, the longer and harder the trip, the higher the bill.

Because of these costs, some AI companies are reconsidering simple pricing plans like flat monthly fees. Instead, they are exploring pricing that better matches how much a customer uses the AI. A common approach is charging by “tokens”, which are small pieces of text that the AI reads and writes (like counting words, but in smaller chunks).

This shift is happening as major AI labs, meaning the companies building the most advanced models, move closer to possible initial public offerings, or IPOs. An IPO is when a private company starts selling shares to the public on the stock market. Ahead of that, investors tend to look closely at whether a company’s pricing can cover its costs over time.

What to watch

Watch for more AI tools to show clearer usage limits, or to add pay as you go options alongside subscriptions. Also watch whether companies can lower the cost of running AI enough to keep prices stable. If they cannot, everyday users may see higher bills for premium AI features, and businesses may become more selective about when they use AI.

Source: Financial Times