Amazon’s CEO says their AI tool has saved them an insane amount of time

The result was a “game changer,” Jassy said.

“The average time to upgrade an application to Java 17 has dropped from 50 developer days to just a few hours,” he wrote. “We estimate that this has saved us the equivalent of 4,500 years of development work (yes, that number is crazy, but real).”

The AI ​​is not only fast, but also seems pretty accurate, according to his post. Amazon developers delivered 79 percent of AI-generated code reviews without further changes, Jassy wrote.

“The benefits go beyond how much effort we’ve saved developers,” he said. “The upgrades have increased security and reduced infrastructure costs, delivering an estimated $260 million in annual efficiency gains.”

Jassy’s comments seem to tie into an often used argument when it comes to AI: that it helps free up time otherwise used for boring but necessary jobs.

“One of the most tedious (yet critical) tasks for software development teams is updating core software,” he wrote. “This work is either dreaded or put off for more interesting work—or both.”

But while Amazon may enjoy the increased productivity, developers may be concerned about its efficiency that could reduce the need for human workers.

Amazon Web Services CEO Matt Garman recently said in an internal meeting that software engineers may need to develop other skills as AI proliferates in coding.

“If you go forward 24 months from now, or in a certain period of time — I can’t predict exactly where it is — most developers may not be coding,” he said.

And Garman is not the only executive to voice this warning. Nvidia CEO Jensen Huang said that “everyone is a programmer now” because of AI coding tools. Emad Mostaque, the former CEO of Stability AI, even predicted that “there will be no programmers in five years.”

Developers aren’t the only industry affected by a potential extinction scare. Klarna CEO Sebastian Siemiatkowski said in a now-deleted tweet that the fintech firm will save $10 million this year by using generative AI to do the marketing work previously done by human employees.

“We spend less on photographers, image banks and marketing agencies,” he wrote. “Our in-house marketing team is half the size of last year, but producing MORE!”

Although Siemiatkowski received a strong backlash, a press release backed up its enthusiasm for AI, saying that Klarna “generated over 1,000 images in the first three months of 2024 using genAI, reducing the image development cycle from 6 weeks to just 7 days.”

CMO Michael Lacorazza previously told Business Insider that the company was able to use generative AI to “cut about two and a half months” off the development cycle for a new brand campaign.

Despite the impressive efficiency gains, Lacorazza made it clear to reassure workers that he sees generative AI “not a replacement for humanity, but an enabler for humanity.”

Meanwhile, Jassy said Amazon will continue to use Amazon Q in other operations.

He said, “Not only do our Amazon teams plan to make more use of this transformation capability, but our Q team plans to add more transformations for developers to use.”