International Business Machines Corp. (NYSE:IBM) Chairman and CEO Arvind Krishna questioned enterprise spending on frontier artificial intelligence models, stating that open-weight alternatives now offer nearly equal performance at a fraction of the price.
CEO Asks ‘Why Pay 10x More?’
Speaking during the company’s third-quarter 2026 earnings call, Krishna outlined the shrinking performance gap between the industry’s most advanced frontier large language models and alternative open models. Krishna stated that open-weight models now run at 10% to 25% of the token cost of frontier models.
“Our computation is that for an average large enterprise, you can get going with maybe two or three racks, which is not a huge amount of infrastructure, but it gives you the ability to run those open-weight models. Currently, anywhere from 25% to 10% of the token cost of the frontier models. That is very attractive given that they’re now 90% as effective,” he said.
Krishna noted that in late 2024, open models scored significantly lower on benchmarks than frontier models, but the gap has narrowed to where alternate models are now about 90% as effective.
“It’s a simple question for any enterprise,” Krishna stated. “Would you like something that’s 10% of the cost if it’s 90% as good as the best, or do you want to pay that 10 times higher cost?”
Sovereign AI and Data Control
Krishna said the economics of open-weight models are converging with another major enterprise trend: sovereign AI. Companies increasingly want to keep sensitive data, logs and model activity under their own control, fueling demand for on-premises and private AI infrastructure.
For most daily enterprise tasks, such as summarizing emails, answering customer calls, or examining contracts, Krishna stated these cheaper alternative models are sufficient.
Q3 Earnings and Mainframe Rebound
Krishna’s remarks accompanied IBM’s third-quarter financial results, where the company reported $17.1 billion in total revenue. IBM Senior Vice President and Chief Financial Officer Jim Kavanaugh reported a 9% acceleration in software revenue.
Additionally, Kavanaugh noted a 10% growth in transaction processing, driven by the company’s mainframe software.
Validating management’s earlier claims that second-quarter mainframe softness was merely a timing issue with slipped deals, technology analyst Patrick Moorhead posted on X that the previous quarter’s drop was an “aberration.”
How Has IBM Performed in 2026?
The stock was 4.64% lower in overnight trading. It has declined 10.85% over the last year, down 15.11% year-to-date, but up 8.42% over the last month. The stock closed up 3.94% at $251.464 on Wednesday.
Benzinga’s Edge Stock Rankings indicate that IBM maintains a strong price trend in the short term but a weak trend in the long and medium terms, with a poor value score.
Disclaimer: This content was partially produced with the help of AI tools and was reviewed and published by Benzinga editors.
Photo courtesy: Shutterstock
Recent Comments