The AI revolution has entered a new phase, one where the all-you-can-eat buffet is over and companies are being forced to count their AI calories. The once-unfettered enthusiasm for AI tools has given way to a more measured approach, with executives and developers rethinking their strategies and imposing limits. This shift is not just a result of rising prices; it's a broader cultural change, a reckoning moment for many companies and a sign that the AI bubble may be bursting.
The era of tokenmaxxing, where companies spent freely on AI tools without much consideration for cost, is over. Now, with prices rising and budgets tightening, companies are being forced to think more strategically about their AI spending. This is particularly evident in the tech industry, where companies like Coinbase, Salesforce, and Uber have had to implement sophisticated systems to manage their AI spending.
Coinbase, for example, has instituted a system of weekly price caps based on job level and role, while Salesforce has fully opened the floodgates for spending on Anthropic tools but is now trying to find a balance that doesn't divert too much money to the rising startup. These companies are not alone; many others are also rethinking their AI strategies and imposing limits on token use.
The shift towards more measured AI spending is not just a result of rising prices; it's also a sign that companies are becoming more aware of the potential for token waste and the need to focus on value. This is particularly evident in the case of GitHub, which has shifted to a usage-based pricing model, causing some developers to burn through their new monthly quotas quickly.
The new era of AI spending is also marked by a greater focus on cost-cutting measures. Companies are now more likely to offload basic work to less advanced AI models, either from American companies or from Chinese firms like Deepseek and MiniMax. This shift is not just about saving money; it's also about finding the right balance between using the most advanced models and those that are more cost-effective for certain tasks.
The AI coding craze is still in its infancy, but the twists and turns of 2026 have already shown that the era of unfettered AI spending is over. Companies are now being forced to count their AI calories, and the reckoning moment for many is here. The question remains: how will this shift impact the future of AI and the companies that are at the forefront of this revolution?
