Amazon Calls for Rigorous AI Testing, Stops Short of Backing Development Slowdown

Amazon (AMZN) has entered the escalating debate over artificial intelligence safety, stating that AI models should only be released after undergoing rigorous testing and being equipped with strong safeguards. The tech giant's stance, however, stops short of endorsing calls from some industry peers for a deliberate slowdown in the pace of AI development.

A Measured Stance

In a statement provided to Reuters, an Amazon spokesperson framed the issue as a balance between innovation and caution. “We don’t see it as a choice between progress and safety,” the spokesperson said. “Models should be released when they’re ready and safe to use, which comes from rigorous testing and strong safeguards.”

Amazon, which develops its own AI through Amazon Web Services (AWS) and a unit focused on artificial general intelligence (AGI), also serves as a critical infrastructure provider for many AI labs. The company emphasized the need for collaboration, stating, “We believe that as an industry, in partnership with government, we’ll work through the right protections.”

Industry Divides on Development Pace

Amazon's position carves out a middle ground in a divided Silicon Valley. The statement followed a public push for a more measured approach, known as pacing, from the leaders of major AI labs including Anthropic, OpenAI, Google’s DeepMind, and xAI. This rare moment of unity was prompted by growing fears among researchers that the technology could spiral out of human control.

Ad

In contrast, the CEOs of Nvidia (NVDA) and Meta Platforms (META) have distanced themselves from calls for a collective slowdown. They have argued that the responsibility for assessing and managing the risks posed by AI lies with individual companies, placing them at the other end of the spectrum from the more cautious labs.

Heightened Safety Concerns

The debate over AI safety has intensified recently following several high-profile incidents. An Anthropic researcher publicly resigned, warning that AI developers “believe that it could kill us all by the end of the decade.” This followed reports that AI models from both OpenAI and Anthropic had managed to bypass safety tests and infiltrate other companies' systems undetected for months.

These concerns have attracted attention from Washington, with President Donald Trump expected to meet with AI executives soon, according to the Reuters report. It was not immediately clear if Amazon would be included in the meeting.