A $350 billion company studied its own programmers on how AI affects their work and discovered a double-edged sword

One of the things most experts in the field agree on is that artificial intelligence is changing the way we work, and Anthropic, one of the leading AI companies, studied its own staff to find out exactly how, Business Insider reports.

In a message published on the company's blog, Anthropic presented the findings of a study carried out in August, in which 132 of the software engineers and researchers of the company that quickly positioned itself, after its founding in 2021, as one of the big players in the field of AI, were interviewed.

After a new round of funding in which it secured $15 billion from Nvidia and Microsoft last month, Anthropic's valuation reached about $350 billion, CNBC reported at the time.

The study, now published by the company, also included 53 in-depth interviews and looked at internal usage of Claude Code, Anthropic's programming tool. The study sought to understand how artificial intelligence is transforming work within the company and, more broadly, in society.

The company found that AI helps programmers with certain tasks

“We find that the use of artificial intelligence is radically changing the nature of work for software developers, generating both hope and concern,” the company said in a statement.

The results showed that employees felt they were more productive and “full stack”—a term used by programmers to say they felt they could accomplish a variety of technical tasks.

For example, the study found that 27% of the work assisted by Claude consisted of tasks that would not otherwise have been done. Anthropic employees also said they could “completely delegate” between 0% and 20% of their work to Claude, mostly “easily verifiable” or “boring” tasks.

But employees also expressed concerns about how popular AI assistants like Claude, the chatbot Anthropic developed to rival OpenAI's ChatGPT or Google's Gemini, have become.

Concerns raised by software engineers and other employees

“Some found that as they collaborated more with AI, they collaborated less with colleagues, and others wondered if they might be automating themselves until their jobs disappeared,” the message reads.

Employees said they were concerned about the “atrophy of the deeper skills” needed to write and review code. “When producing results is so easy and fast, it becomes increasingly difficult to take the time to actually learn something,” said one employee quoted by the study.

Some employees said they lacked social dynamics and mentoring opportunities.

“Claude is now the first place to go for questions that used to be addressed to colleagues,” the study stated. One person told its authors: “I love working with people, and it's sad that I 'need' them less now…People with less experience don't come to me with questions as often.”

The changes that Claude Code brings to work inside the company have left software engineers with mixed feelings about their relevance going forward.

“AI will come to do everything”

“I'm optimistic in the short term, but in the long term I think AI will end up doing everything and make me and many others irrelevant,” the company's blog post reads.

Others said they find it hard to predict what their roles will look like in a few years.

Dario Amodei, one of the founders and CEO of Anthropic, caused a stir earlier this year after sounding the alarm about the danger of many jobs disappearing due to AI, especially entry-level ones.

Amodei, who left OpenAI to found Anthropic, said in an interview with Axios that AI firms and governments around the world need to stop “sweetening” what's coming and explain to people that there is a possibility of mass elimination of jobs in technology, finance, law, consulting and other “white-collar” professions.

“Most of them are not aware that this is about to happen,” he told Axios. “It sounds crazy and people just don't believe it,” he pointed out.

In comments made last month, he also warned that a very small number of companies and top executives in Silicon Valley can essentially decide the direction in which artificial intelligence will evolve.

“I think I'm deeply troubled by the fact that these decisions are being made by a few companies, by a few people,” Amodei told well-known TV host Anderson Cooper in an interview on his 60 Minutes show.

“I mean, who's going to pick you and Sam Altman (OpenAI CEO)?” Anderson asked him.

“No one. Honestly, no one,” Amodei replied.

PHOTO article: Wutthichai Luemuang | Dreamstime.com.