American companies are trying to figure out how to rein in increased spending on artificial intelligence. That’s helping payments software giant Lamp.
The spend management company announced a $750 million funding round Thursday at a valuation of $44 billion. The round was led by ICONIQ, GIC and Ontario Teachers’ Pension Plan, increasing Ramp’s valuation by approximately 38%.
CEO Eric Greiman said the New York-based company had positive free cash flow and annual revenue of more than $1 billion. Some of this growth is coming from enterprise customers who are grappling with AI spending that is eating up a large portion of their budgets.
“What we know is that tokens come at a significant cost, and most CFOs not only didn’t plan for this, exponential growth, in their annual plans, they don’t have good tools to manage this,” Greiman said in an interview with CNBC on Thursday. “Suddenly there’s a third pillar of spending through tokens and intelligence, which is not a pretty spending area.”
Eric Greiman and Karim Atiyeh, co-founders of corporate card startup Ramp
Ramp currently offers products to help clients manage their AI spending. This allows businesses to route tasks to AI models and execute them at a fraction of the cost. That compensation for CFOs often comes in the form of payments in “tokens,” the units that AI companies use to measure usage.
Griman said CFOs are often surprised by how much they’re actually spending.
“People are saying this is the biggest opportunity to grow the business in our careers, and yet this is the fastest growing item,” he said. “The problem is that most companies use the frontier model, this cutting-edge intelligence for everything. … Don’t get me wrong, you might need super-advanced intelligence to perform the most important analysis, but you might not need it to edit an email.”
There is also the question of return on that expenditure.
Griman said companies spending the most on AI are seeing the biggest increase in revenue, with some achieving “extraordinary ROI.” But the benefits often come from companies investing effectively in AI.
Griman said that of the 70,000 companies using Ramp, those that spend the majority of their revenue on AI saw a 12% increase in revenue. Companies with the lowest spending saw flat growth.
So far, that spending hasn’t come at the expense of software budgets.
“Despite what’s happened in the stock market, we haven’t seen any software spending yet,” Greiman said. “The growth continues, but I think the bill is due.”
Gilman said frontier model companies like OpenAI and Anthropic have no reason to steer people toward cheaper options.
“They have no incentive to tell someone who knows the job you wanted to do. It can be done at 1/100th the cost,” he says. “Your incentive is really to maximize revenue and profit. So I think that’s what’s leading to the rise of both companies like Ramp that can help companies bring in these token spends and checks, and AI-native companies that are making decisions to provide the resources to route tasks to the most cost-effective options.”
He also mentioned “tokenmaxxing,” an approach where developers use as many tokens as possible. Some companies use it as a proxy for productivity, but the problem is that more tokens doesn’t necessarily mean more value.
“I think we are now in the twilight of token max,” he said, adding that companies are smartly aware of that metric. “I think the era of token maxing is coming to an end.”
