A New Security Startup Says AI Chips Have a Blind Spot. It Wants to Fix That.

Key points

• Stealthium is a new cybersecurity startup focused on detecting attacks inside AI accelerators, the specialised chips used to run artificial intelligence workloads.
• Current security tools largely cannot monitor activity happening inside these chips, leaving a gap that criminals could exploit.
• The company targets "neo-clouds", smaller cloud computing providers built specifically to rent out AI processing power.
• Stealthium's approach analyses subtle behavioural signals from hardware to flag unusual activity that conventional tools miss.

Most companies running artificial intelligence have no idea what is happening inside the chips that power it. That is the gap Stealthium, a newly launched security startup, is trying to close.

The chips in question are called AI accelerators, which are specialised processors designed to handle the heavy maths that AI models require. Graphics processing units, or GPUs, are the best-known type. They sit at the heart of almost every major AI service today, from chatbots to image generators, and they process enormous amounts of sensitive data. Until now, mainstream security software has focused on monitoring servers, networks, and software, and has largely left what happens inside these chips invisible.

Stealthium's pitch is that this blind spot is real and exploitable. The company says it watches for subtle telemetry signals, meaning tiny streams of performance and behaviour data that the chip emits while working, and uses those signals to flag patterns that do not fit normal operation. Think of it like listening to a car engine for sounds that suggest something is about to go wrong, rather than waiting for the warning light.

Who is most at risk?

The companies with the most exposure right now are what Stealthium calls "neo-clouds", smaller cloud computing providers whose entire business is renting out AI chip power to customers. Traditional cloud giants have large security teams and years of hardened infrastructure behind them. Neo-clouds are newer, growing fast, and often running hardware at scale before security processes have caught up.

Any business renting AI computing time from one of these providers, or running its own GPU cluster for AI development, sits in the same exposed position. If an attacker found a way to abuse activity inside the accelerator hardware, conventional security tools would likely miss it entirely.

No specific breach or CVE, meaning a numbered record of a publicly known software flaw, has been attributed to this class of weakness yet, at least not publicly. Stealthium's argument is that the absence of known incidents does not mean the risk is absent. It may simply mean nobody has been watching.

SecurityWeek first covered the company's launch and its focus on this hardware-level monitoring gap.

What should ordinary businesses do right now?

If your organisation rents AI computing power from any provider, ask that provider directly what security monitoring they have in place at the hardware level. Most will not have a satisfying answer yet, and that is useful information.

For teams running their own GPU infrastructure, review whether your security monitoring reaches the hardware layer, or stops at the operating system. Awareness of where your visibility ends is the first practical step.