Nvidia has denied allegations of security vulnerabilities in its AI processors following an inquiry by the Cyberspace Administration of China (CAC) into its H20 GPUs.

The US-based semiconductor company said in a blog post on August 5, 2025, that its chips do not contain back doors, kill switches, or spyware.

Nvidia’s chief security officer, David Reber Jnr, wrote that embedding weaknesses in chips would harm global digital infrastructure and damage trust in US technology.

The company said its graphics processing units are used in sectors such as healthcare, finance, and scientific research.

Nvidia said that existing laws require companies to address vulnerabilities, not create them.

A company representative also told the South China Morning Post that its chips do not offer remote access or control to any party.

The CAC had summoned Nvidia last week to discuss security concerns related to its products.

🔗 Source: South China Morning Post

🧠 Food for thought

1️⃣ Hardware security concerns follow decades-long pattern of cyber espionage targeting critical infrastructure

China’s demand for security evidence from Nvidia reflects a broader historical pattern of nation-state concerns about embedded vulnerabilities in critical technology infrastructure.

Major cyber espionage cases over the past 25 years show how hardware and software systems have repeatedly been exploited by hostile actors, from the Moonlight Maze attacks on U.S. military networks in 1998-1999 that stole thousands of military technology documents, to Operation Aurora in 2010 where Chinese hackers compromised Google and 20+ other companies1.

The OPM data breach from 2012-2015 demonstrated how vulnerabilities can lead to massive intelligence losses, with Chinese hackers stealing personal data of 21 million Americans including sensitive federal employee information1.

China’s specific concerns about Nvidia’s H20 chips potentially containing “tracking and remote shutdown capabilities” echo these historical precedents where embedded vulnerabilities created systemic risks2.

The Chinese regulatory summons follows scenarios outlined in state media about catastrophic failures from chip vulnerabilities, such as power loss in vehicles or failures during remote surgeries, showing how these security concerns extend far beyond simple data protection2.

2️⃣ Semiconductors have become the new battleground for technological sovereignty