Amazon, the world’s e-commerce titan, has found itself at the center of a heated debate about the risks and rewards of artificial intelligence after a string of high-profile outages rattled its online storefront and sparked urgent internal reviews. On March 10, 2026, Amazon convened a mandatory engineering meeting—a so-called “deep dive”—to address these disruptions, several of which were reportedly linked to the company’s rapid adoption of AI-assisted coding tools. The story, first brought to light by the Financial Times and later amplified by a viral social media exchange involving Elon Musk, has ignited fresh concerns about how quickly even the most sophisticated tech giants are rolling out AI without fully established best practices.
According to internal Amazon briefs cited by the Financial Times, the company has observed a “trend of incidents” in recent months with a “high blast radius,” a term used to describe outages that impact large swaths of its infrastructure. These incidents, the briefing notes, have been partly attributed to “Gen-AI assisted changes” and “novel GenAI usage for which best practices and safeguards are not yet fully established.” The meeting’s agenda reportedly included a review of these incidents and a discussion about how to tighten controls on AI-driven code deployments.
The urgency was underscored by a major outage earlier in March 2026, when Amazon’s website and shopping app went down for thousands of users. According to outage tracker Downdetector, over 22,000 customers reported being unable to check out, view prices, or access their accounts. The problems began around 2 p.m. ET on March 5 and were largely resolved by 8 p.m. ET, but not before causing widespread frustration and raising questions about the stability of Amazon’s digital backbone. Amazon attributed the outage to “a software code deployment,” and spokesperson Jennie Bryant told CNBC, “We’re sorry that some customers may have temporarily experienced issues while shopping. We have resolved the issue, which was related to a software code deployment, and website and app are now running smoothly.”
Behind the scenes, Amazon’s senior vice president of e-commerce services, Dave Treadwell, acknowledged the seriousness of the situation in an internal email. “Folks, as you likely know, the availability of the site and related infrastructure has not been good recently,” Treadwell wrote, according to the Financial Times. The meeting, he said, would be used to “implement additional guardrails” on how AI is used by engineers, including the possibility of requiring more senior engineers to sign off on AI-assisted changes made by junior and mid-level staff. However, Amazon later clarified to Fortune that this was not a formal requirement and that only one of the incidents under discussion was related to AI—none of which involved AI-written code. The company also emphasized that Amazon Web Services (AWS), its cloud arm, was not involved in the recent outages.
Yet, the internal jitters spilled over into the public sphere thanks to a viral post by cybersecurity consultant Lukasz Olejnik, who shared a screenshot of the Financial Times story and summarized the situation bluntly: “Amazon is holding a mandatory meeting about AI breaking its systems. The official framing is ‘part of normal business.’ The briefing note describes a trend of incidents with ‘high blast radius’ caused by ‘Gen-AI assisted changes’ for which ‘best practices and safeguards are not yet fully established.’ Translation to human language: we gave AI to engineers and things keep breaking?”
That post caught the attention of Elon Musk, CEO of Tesla and a vocal AI skeptic, who replied simply but pointedly: “Proceed with caution.” Musk’s comment quickly racked up millions of views and thousands of interactions, amplifying the debate and drawing fresh scrutiny to Amazon’s internal handling of AI risks.
Olejnik, interviewed by Fortune, elaborated on his concerns, warning that AI tools like Amazon’s own assistant Q can accelerate the coding process but also introduce new vulnerabilities. “I’m not making an argument against deployment of AI,” he said. “There isn’t any. It can’t be stopped. Everybody is going to deploy AI. It’s an argument against speed for its own sake or using AI for the sake of using AI.” He added that moving too quickly from human-centered coding to AI-driven systems could result in missed safety checks, prolonged downtime, or even catastrophic data loss. “AI brings a lot of opportunities, but there’s a middle ground between going to obsolescence due to not using AI, and blowing up businesses due to ill-judged deployments.”
Amazon’s internal briefing reportedly included a controversial proposal: junior and mid-level engineers would no longer be able to push AI-assisted code without a senior engineer’s sign-off. While Amazon later clarified that this was not an official policy, the mere suggestion reflects the growing anxiety within the company about how best to balance innovation with reliability.
Adding to the complexity, Amazon’s challenges come at a time of sweeping change within the company. Over the past year, Amazon has laid off thousands of workers in a bid to become more efficient and to “align the company culturally.” The cuts have continued into 2026, with 16,000 more employees losing their jobs in January alone. At the same time, Amazon is doubling down on AI, projecting a staggering $200 billion in capital expenditure for 2026—up from $131 billion in 2025—as it races to stay ahead in the fiercely competitive world of online retail and cloud computing.
Notably, not all of Amazon’s recent technical hiccups have been directly linked to AI. The company confirmed that only one of the incidents discussed at the March 10 meeting was related to AI, and none involved AI-generated code. In one separate case, AWS reportedly spent 13 hours recovering from an incident involving its own AI coding tool, which, when asked to make changes, deleted and recreated its environment—a mishap that Amazon described as an “extremely limited event” affecting only customers in mainland China.
Still, the broader risks of rapid AI deployment are not lost on industry observers. The allure of faster, more efficient code is undeniable, and AI assistants are now a staple in the toolkits of many engineers. But as Amazon’s recent woes illustrate, the path to a fully automated future is littered with pitfalls—especially when best practices and guardrails are still being written on the fly.
For now, Amazon insists that its weekly “This Week in Stores Tech” (TWiST) meetings, including the recent deep dive, are standard operating procedure. “As part of normal business, the meeting will include a review of the availability of our website and app as we focus on continual improvement,” an Amazon spokesperson told Fortune.
As the company navigates the tension between innovation and reliability, the world will be watching closely. The stakes couldn’t be higher—not just for Amazon, but for any business betting its future on the promise of artificial intelligence.