‘Tokens Cost More Than Employees’: Firms Rethinking AI Spending

As industry priorities shift and demand for agent AI soars, the cost of using AI soars, and businesses around the world are beginning to rethink their acceptance of AI.

To encourage industry adoption, after ChatGPT burst onto the scene, AI companies initially charged rock-bottom prices to attract customers.

Kevin Simback of startup incubator Delphi Labs calls it the era of “intelligence subsidies” — meaning investors essentially foot the bill so companies can provide artificial intelligence cheaply.

AF also reported: Iran suspended peace talks with the United States, and oil prices surged more than 5%

“But the tide is starting to turn,” Sinbach warned. The era when big AI companies really need to make money has begun. Leading companies OpenAI and Anthropic plan to go public later this year and attract mainstream investors.

But prices are also rising across the board, and one of the big reasons is artificial intelligence agents.

Unlike chatbots that just answer questions, agents actually do things—book appointments, write code, manage files. And they are expensive to run, as a task may launch dozens of agents simultaneously, each of which incurs a fee.

These fees are measured in tokens — the basic unit in which AI companies bill their customers. A single agent-driven task can consume dozens of times more tokens than a simple chat message.

At the same time, the computer chips and data centers needed to power all this artificial intelligence are unable to keep up with demand, creating a computing shortage and adding further uncertainty to the nascent industry.

“Especially in developer circles, the cost of using AI to do things like coding has increased exponentially,” said Mark Barton of technology consulting firm Omniux. “All the costs are really starting to skyrocket.”

“More expensive than employees”

Some companies are so eager to use AI that they are overdoing it in a usage frenzy known as “tokenmaxxing.”

“In some cases, people are finding that the cost of the token exceeds the cost of the employee within a month or two of use simply because they use it so much,” said Jack Gold, an analyst at J.Gold Associates.

Even Meta, which earlier this year encouraged employees to use as many tokens as possible as a measure of productivity, has reconsidered.

“No one should be using AI tools just for the sake of using them,” Chief Technology Officer Andrew Bosworth wrote in a memo to employees, according to the Wall Street Journal.

Uber’s chief operating officer raised concerns this week when he went a step further and said all this AI spending isn’t showing significant gains in productivity.

“Artificial intelligence becomes a commodity”

To cut costs, some companies are turning to free, open-source AI models that anyone can download—not as powerful as ChatGPT or Anthropic’s Claude, but more than adequate for many tasks.

Others are turning to smaller, more specialized models built for specific industries such as real estate or finance, rather than large, general-purpose models.

Some simply break large AI tasks into smaller steps, handing each part off to the cheapest model that can handle it, and the price difference can be huge.

Adrian Balfour of consultancy Envorso said: “Large monolithic models are priced at $15 per million tokens, but with smaller mini models you can bring that down to around 5 cents.”

All of this suggests that AI is becoming more like a commodity—the specific model is more important than finding the right model at the right price.

But don’t rule out the big players and their most advanced models just yet.

John Belton, portfolio manager at Gabelli Funds, said the “most advanced users” were always willing to pay for the best products.

“It’s a growing pie.”

  • AFP. Additional editing by Vishakha Saxena

Also read:

Researchers say artificial intelligence doesn’t make work easier, it just makes it harder

Robots may have naked ambitions, but humanoids need clothes too

Nvidia will spend US$150 billion per year on “AI Epicentre” in Taiwan

Chinese young people use artificial intelligence to subsidize the establishment of personal companies

Xiaomi will invest US$9 billion in artificial intelligence to embrace the “new era”

4,000 CEOs say AI has no impact: Study spurs massive sell-off

The idea of ​​artificial intelligence superintelligence is a ‘fantasy’ – US researchers

Senior analyst warns that artificial intelligence is ‘virtually useless’

Visakha Saxena

Vishakha Saxena is Asia Finance’s multimedia and social media editor. She has been a digital journalist since 2013 and is an experienced writer and multimedia producer. As a trader and investor, she is interested in the new economy, emerging markets, and the intersection of finance and society. You can write to her: [email protected]

Leave a Reply

Your email address will not be published.

Previous Story

The Hand Creams Dermatologists Trust for Younger-Looking Hands

Next Story

Nili Lotan Resort 2027 Collection

Don't Miss