The chipmaker responded bluntly after Chinese regulators raised concerns about the integrity of Nvidia’s hardware, and as some US lawmakers float legislation that could require exactly that.

The company said: “There are no back doors in Nvidia chips. No kill switches. No spyware. That’s not how trustworthy systems are built and never will be.”

The firm published a blog post explicitly distancing itself from the idea, criticising what it called “pundits and policymakers” for pushing proposals that would actively weaken digital infrastructure. Nvidia argued that such measures would hand hackers and hostile actors easy access to critical systems and fracture global trust in US-made technology.

Nvidia added that embedding kill switches would echo the failed “Clipper Chip” programme in the 1990s when the US government tried to bake backdoors into encryption hardware. That ended in a security mess, with attackers exploiting the system and leaving American networks more vulnerable.

“Established law wisely requires companies to fix vulnerabilities not create them,” said Nvidia.

It is not a coincidence that Nvidia’s stance arrived just as Chinese regulators began scrutinising the firm’s AI chip shipments, especially the H20, which has faced supply issues thanks to US export controls. The timing suggests Nvidia is trying to reassure both Beijing and Washington that its chips can be trusted without adding legally mandated sabotage points.

Nvidia may also be looking to head off a future Senate bill that could force chipmakers to include security override features. So far, nothing has been passed but the conversation is heating up, especially as AI hardware becomes strategically vital.