Explore our new Cybersecurity Terms Glossary, explained by our experts.
Cybersecurity Terms Glossary
A hardware Trojan is a malicious modification made to an integrated circuit or board during design or manufacture, dormant until a rare trigger activates a payload that leaks data or disables the device.

What Is a Hardware Trojan?
Software can be rebuilt from source; silicon cannot. A hardware Trojan is a deliberate malicious modification introduced into an integrated circuit or a printed circuit board at some point between specification and delivery. It sits dormant through testing, waits for a condition that ordinary operation will not produce, and then leaks keys, degrades performance, grants privilege, or simply stops the device working.
Academic work on the subject, much of it indexed through the Trust-HUB taxonomy, splits a Trojan into two parts. The trigger decides when to act; the payload decides what to do.
A good trigger is one that normal validation will never produce. Counters that fire after millions of cycles, a specific input sequence an attacker can supply, an on-chip sensor reaching an unusual temperature or voltage, or a particular plaintext arriving at a cryptographic block. Because functional testing exercises only a tiny fraction of a modern chip’s state space, a rare trigger is effectively invisible to it.
Payloads tend to fall into four groups: leaking secrets through a side channel or an unused pin, weakening a random number generator so that keys become predictable, granting a privilege escalation path, or acting as a kill switch that disables the part on command. The third group is the most dangerous commercially and the fourth is the most discussed in defence contexts.
The modern chip supply chain is long enough to offer several opportunities. Third-party intellectual property blocks are licensed and integrated without the buyer seeing the source in every case. Design and layout tools can be subverted. Fabrication is overwhelmingly outsourced to foundries that receive the full design. Packaging, test, and distribution follow, and counterfeit or remarked components re-enter the chain through brokers.
The most quoted research in this area is the A2 attack, presented at IEEE Security and Privacy in 2016, which built an analog trigger out of a single additional capacitor-like structure that charged up only when a specific instruction sequence ran repeatedly. It occupied a negligible area, passed functional testing, and escalated privilege when triggered. The point of the work was that a Trojan need not be a digital circuit that stands out in a netlist comparison.
Claims of Trojans in commercial products should be treated carefully. The 2018 Bloomberg report alleging implanted chips on Supermicro server boards was denied by the named companies and has never been substantiated publicly, and it remains a case study in how hard verification is rather than a confirmed incident.
Design-time countermeasures therefore matter more than inspection. Split manufacturing divides a design across foundries so that no single one sees the whole thing. Logic locking requires a correct key for the circuit to function. Built-in self-authentication and on-chip monitors watch for behaviour outside specification. Formal equivalence checking between register transfer level and gate-level netlists catches digital insertions during synthesis.
Unless you are designing silicon, your exposure is procurement rather than fabrication. Buy from authorised distributors rather than brokers, because counterfeit and remarked parts are a far more likely problem than a nation-state implant. Keep a hardware bill of materials with component provenance for anything in a critical system. Require suppliers to describe their own supply chain assurance, and design systems so that a single untrusted component cannot compromise the whole, by placing keys in a separate trusted platform module and validating outputs independently.
The NCSC supply chain guidance is the reference point for that programme, and for telecommunications operators the Telecommunications (Security) Act 2021 and its code of practice set specific expectations about equipment provenance and vendor assurance. Our Hardware Penetration Test Services examine devices at board level, which is where realistic tampering is found.
Confirmed public cases in commercial silicon are vanishingly rare. Most documented examples come from academic research and from counterfeit or remarked components, which are common. The widely reported 2018 allegation about implanted chips on server motherboards was denied by the companies involved and never substantiated. That does not make the risk theoretical for defence and critical national infrastructure, but it should be kept in proportion.
Not from a fabricated chip. The modification is part of the physical device, so there is no patch, no reflash, and no configuration change that removes it. The response is replacement of the affected hardware, which is why prevention through design-time controls, trusted foundries, and procurement assurance carries so much more weight here than in software security.
A firmware implant is malicious code in rewritable flash memory, so it can be detected by reading the image and removed by reflashing a verified version. A hardware Trojan is in the silicon or on the board itself. Firmware implants are far more common and far more relevant to most organisations, which is why firmware inventory and update control deserve attention first.
Ask where components are sourced and whether distribution is authorised, how long firmware and hardware support will be provided, what the vulnerability disclosure process is, and whether a hardware bill of materials can be supplied for critical systems. For telecoms operators in the UK, the Telecommunications (Security) Act code of practice sets out specific vendor assurance expectations worth reusing as a procurement template.
You May Be Interested In These
We use a few cookies that are needed to make this website work, and they are always on. We would also like to set analytics cookies, to see which pages are useful; functional cookies, so that our live chat works; and advertising cookies, to measure our campaigns. We will not set any of those until you say yes, and you can change your mind at any time from the Cookie Settings link in the footer.
Choose the cookies you are happy for us to use. Strictly necessary cookies are always on; everything else is off until you turn it on.
These cookies keep the website working: they hold your security checks, your form submissions and the choice you make here. The site cannot run without them, so they are always on.
These let our live chat open and remember your conversation, and let the newsletter sign-up work. Switch them off and the chat window will not appear.
These tell us which pages are read and how people find us, so that we can improve the site. The reports are about groups of visitors, never about you personally.
These let us measure our advertising and show you our adverts on other websites, such as LinkedIn, Google and Meta. Switch them off and you will still see adverts, but they will be less relevant.