AI technology is becoming more expensive, leading some companies to rethink their use of it. Initially, AI firms like OpenAI and Anthropic offered services at very low prices to attract customers. Kevin Simback from Delphi Labs calls this “subsidized intelligence,” where investors covered costs to keep prices low. However, with companies looking to go public, the focus is now shifting to profitability.
The rising costs are significant. Unlike chatbots that only respond to queries, AI agents perform tasks like scheduling appointments or writing code. This can lead to heavy usage of resources, as a single task may require multiple agents, driving up expenses. Charged in tokens, the costs can escalate quickly—sometimes dozens of times more than a simple chat might incur.
A growing demand for computing power is creating shortages. Mark Barton from Omniux points out that costs for using AI in programming have spiked. Some companies have overextended their AI usage, a trend referred to as “tokenmaxxing.” Analyst Jack Gold warns that in some cases, the cost of AI tokens can exceed what would be paid for an employee.
Even big tech firms are re-evaluating their strategies. Meta’s CTO Andrew Bosworth advised against using AI just for the sake of it. Similarly, Uber’s COO highlighted that big spending on AI hasn’t led to noticeable productivity gains.
In response, companies are increasingly turning to free, open-source AI models. These alternatives may not be as powerful as options like ChatGPT but can still handle many tasks effectively. Some firms are opting for smaller models tailored to specific industries, cutting down costs significantly.
Adrian Balfour from Enverso explains that while large models can go for $15 per million tokens, smaller models can be as low as five cents. This shift suggests that AI is becoming a commodity, where the choice of model matters less than finding a cost-effective solution.
Still, premium services won’t be fully replaced. John Belton from Gabelli Funds believes that top users will always invest in the best technology.
As companies navigate the evolving landscape of AI costs and benefits, staying informed on changes in the industry becomes crucial. For more insights on this topic, you can check the Wall Street Journal and explore further analyses on ongoing trends.
Source link
companies, Kevin Simback, Jack Gold
