OpenAI to focus on ‘practical adoption’ in 2026, says finance chief Sarah Friar

OpenAI will make 2026 the year of “practical adoption”, says the AI ​​startup’s finance chief. blog Sunday.

“The priority is to close the gap between what AI currently enables and how people, companies, and countries use it every day,” OpenAI CFO Sarah Friar wrote. “The opportunity is large and urgent, especially in healthcare, science and entrepreneurship, where better intelligence translates directly into better outcomes.”

In the blog, Friar explained how he sees OpenAI’s strategy to increase monetization of services like ChatGPT and secure the computing needed to power these products, and said the AI ​​lab’s revenue tracks directly with the availability of its technical infrastructure. Friar said OpenAI’s calculation increased from 0.2 gigawatts in 2023 to about 1.9 GW in 2025, while the company’s annual revenue rate similarly increased from $2 billion in 2023 to more than $20 billion last year.

“This is growth never seen before on this scale,” he wrote. “And we firmly believe that more computing in these times will lead to faster customer adoption and monetization.”

Friar’s blog comes at a time when OpenAI and the tech industry’s focus on artificial intelligence are coming under scrutiny for the massive investments required to build data centers and secure the energy and components needed to power this cutting-edge technology that hasn’t yet generated much revenue for investors.

These important agreements include the agreement between OpenAI and the AI ​​chip maker Nvidia It was announced in September. Under that agreement, Nvidia said it would commit $100 billion to support its AI initiative while building and deploying at least 10 GW of Nvidia systems.

A gigawatt is a measure of power, and 10 GW is roughly equivalent to the annual energy consumption of 8 million U.S. households, according to a CNBC analysis of data from the Energy Information Administration.

But in November, Nvidia told investors there was “no assurance” that its deal with OpenAI would move beyond an announcement to a formal contract stage.

“Securing world-class computing requires commitments made years in advance, and growth does not follow a perfectly smooth line,” Friar wrote, adding that the system requires discipline.

Three years ago, OpenAI relied on a single computing provider, but now works with a diversified ecosystem, Friar said.

“We can confidently plan, finance and deploy capacity in a market where computing access defines who can scale,” he wrote.

Friar said OpenAI’s business model needs to scale along with its services.

“As intelligence advances into scientific research, drug discovery, energy systems, and financial modelling, new economic models will emerge,” he wrote.

Blog comes after OpenAI last week in question It said it plans to start testing ads for some ChatGPT users in the US. The move comes as OpenAI prepares for a potential public launch this year.

“Making money needs to be experience-specific,” Friar wrote. “If it doesn’t add value, it doesn’t belong.”

WRISTWATCH: OpenAI Investor Letter: Weekly and daily active user figures ‘continue to produce all-time highs’