When artificial intelligence began changing his business four years ago, Aaron Levie, the chief executive of Box, decided to adjust.
Box, a Silicon Valley maker of software for storing and managing data, started weaving A.I. into its products. It began selling A.I. aimed at automating tasks like reviewing and approving contracts. And it asked employees to set aside their usual work to help other colleagues use A.I.
That led Box to post a new job last fall: senior director of A.I., data and integration, to help it wire together its internal systems and data so its workers could better use A.I.
The job was one of 13 new kinds of roles that Box has created because of A.I., with titles like A.I. architect, A.I. solutions manager and A.I. platform leader. With the proliferation of these positions, Box expects to have more than 3,000 employees by early next year, up from 2,900 at the start of this year.
“We ourselves are selling A.I. to our customers, so that’s actually causing us to need to hire more people,” Mr. Levie, 41, said. “And as a user of A.I., we’re getting new forms of productivity that’s also causing us to hire people.”
Across the tech industry, companies such as Meta and Coinbase have been laying off workers in the name of A.I. Many people fear that the powerful technology, which can generate code and answers to questions, will increasingly replace workers in jobs like computer programming and engineering management.
But Box shows that A.I. can also spur job creation, making the labor picture more nuanced. Demand for cybersecurity experts has soared because of an explosion of A.I.-generated code that needs to be vetted, for instance, while companies including Google are recruiting more engineers to help integrate A.I. into their customers’ systems.
The growth of such roles is unlikely to make up for A.I.-related job cuts. With the technology still evolving, it is unclear whether these jobs are permanent or temporary, and whether companies can find workers with the skills to do the work, said Stephan Meier, a professor of business strategy at Columbia Business School.
But new technologies have long created new jobs, he said. In the 1970s and 1980s, the arrival of computers in the workplace led to information technology departments, so “there is now I.T., there is a chief information officer and so on,” Mr. Meier said. “Completely new jobs, new businesses, new degrees were created.”
Mr. Levie helped found Box in 2005. The company’s software, which is often deployed behind the scenes, helps companies store and work on documents and other data. The company has more than 100,000 customers, including federal agencies and Morgan Stanley, and went public in 2015.
Box has typically taken a slow-and-steady approach to hiring. The company, based in Redwood City, Calif., did not go on a hiring binge during the pandemic, as much of the tech industry did. From 2019 to 2022, its work force increased 20 percent to about 2,500 employees. Some of its peers, which have cut thousands of jobs in recent months, doubled or tripled in size during those years.
“We have never done any kind of broad-scale layoffs, and that remains a huge commitment of ours,” said Jessica Swank, Box’s chief people officer.
When OpenAI kicked off the A.I. frenzy with its release of ChatGPT in 2022, Mr. Levie said, he was excited by the technology’s power. Box has infused its software with A.I., adding features where the technology performs tasks like drafting documents.
The moves are starting to pay off. Box’s revenue in the most recent quarter rose 11 percent from a year earlier, the company reported in May, its fastest growth since 2022. But anxieties over whether A.I. will eventually replace Box’s software have weighed on the company, which has seen its stock decline about 7 percent this year.
Mr. Levie has said he believes companies will continue buying software, rather than making their own with A.I., since third-party software is likely to be more secure and reliable. In the meantime, his company is investing in incorporating A.I. into its products.
Box is also investing in the new jobs, which it sees as long-term roles. As A.I. models advance, the way they answer questions or are structured can change, Mr. Levie said. That means Box’s customers will need help using new models, he said.
“It’s kind of a question of, when does A.I. slow down?” Mr. Levie said. “That would be the moment when maybe you have some sort of sustaining, plateauing” of hiring.
Among the jobs that Box is adding are “forward deployed engineers,” who will help customers that may want to use A.I. but do not have the technical know-how. Another is “A.I. business automation engineers,” who are part of the I.T. department and help colleagues use A.I. to be more productive and to remove drudgery from their jobs.
Box also created new roles for evaluating A.I. models, hiring people like Sidharth Srinivasan, 23. Mr. Srinivasan, who joined Box full time last year after graduating from Stanford University, measures the performance of A.I. models to help customers decide which models are best for certain tasks.
“If I talked to myself two years ago, and I told myself I was working on A.I. evals, I would be like, what is that?” Mr. Srinivisan said. “With any technological innovation, the type of work that you have to do to adapt to that is just slightly different.”
The company said its pace of hiring software engineers had not slowed, even as A.I. has improved at coding. With the arrival of A.I. agents that can perform tasks autonomously, a software engineer can manage agents and achieve what once would have required a team of engineers, Mr. Levie said. That means every additional engineer contributes more, he said, so hiring them is even more worthwhile.
“A role that we’re just adding more of is in core engineering,” Mr. Levie said. “Now that we can basically build way more features for our customers, it actually for us is attractive to have more engineers doing that.”
The increased productivity from A.I. has allowed Box to justify hiring people it would not have in the past, Mr. Levie said. The company has recruited employees to market to specific industries, something it could not do before because the tasks would have needed too many workers, he said. But A.I. has made this work more efficient and more possible.
“Now, you’re hiring one or two to do the work of 10 because you can finally afford to do that work,” Mr. Levie said.
