Amazon has entered the debate over artificial intelligence safety, stating that AI labs should only release models after rigorous testing and strong safeguards. The company emphasized that progress and safety are not mutually exclusive, but did not call for a slowdown in AI development.
Amazon's public stance on AI safety adds another voice to the growing industry discussion about responsible AI development, potentially influencing regulatory approaches and the pace of future AI advancements.
Amazon has entered the debate surrounding artificial intelligence safety, stating that AI laboratories should ensure models are released only after rigorous testing and the implementation of strong safeguards. A spokesperson for the Seattle-based firm told Reuters that "We don't see it as a choice between progress and safety," adding that models should be released when they are ready and safe to use.
The company's stance comes days after many of the largest AI labs, including OpenAI, xAI, Google's DeepMind, and Microsoft, called for slower AI development, known as pacing. This rare moment of industry unity followed concerns from researchers that unchecked AI could evade human control and pose threats to global systems and humanity itself. Amazon acknowledged that "There are risks if that doesn't happen collectively," and expressed belief that "as an industry, in partnership with government, we'll work through the right protections."
Amazon develops its own AI models through its Amazon Web Services (AWS) subsidiary and a dedicated unit focused on artificial general intelligence (AGI), which powers services like the voice assistant Alexa. The company also serves as a leading infrastructure provider for AI labs and offers access to competitors' models through its Bedrock marketplace. Concerns over AI safety have reached the White House, though President Donald Trump has dismissed these worries as a "hoax."