Building a competitive moat in the AI era isn’t easy.
AI is advancing so quickly that many companies struggle to plan more than a few months ahead, making it difficult to build a lasting advantage.
OpenAI CFO Sarah Friar has three key tips for companies looking to stand out.
First order of business is for companies to ask themselves whether they’re solving a real problem, she said on CNBC’s Squawk Box on Wednesday. “It sounds really trite, but sometimes I think people try to solve problems that don’t really exist in the world.”
Instead, they should focus on automating complicated tasks that are necessary for real-world operations. There are a lot of complex business processes, especially in areas like finance, that can be automated with agents, she said.
Lastly, she said that companies should focus on securing unique data sets.
“90 plus percent of the world’s data sits behind closed doors,” she said. “It sits in university settings, and company settings, and so on. And so, can you access that in an appropriate way? That’s what I think builds a competitive moat.”
Related stories
Business Insider tells the innovative stories you want to know
Business Insider tells the innovative stories you want to know
The hunt for unique data is so fierce that even leading AI companies are pushing boundaries — sometimes at the risk of copyright violations. Meta considered acquiring publisher Simon & Schuster as a solution. Anthropic collected and scanned millions of pirated books while training its Claude model, which a district judge ruled in June did not constitute fair use.
Finding proprietary data that isn’t publicly available, and getting permission to use it, could be a legally sound foundation for a new AI company.
Friar joined OpenAI last year along with Kevin Weil, now the company’s chief product officer. She previously served as the CEO of Nextdoor and the CFO of Square. She has also worked at Goldman Sachs, McKinsey, and Salesforce.