Ai attacks can hack brain controlled devices and steal thoughts
NERVE Attacks: Breaking AI-Powered Brain-Computer Interfaces
Cryptography and Security
Summary
Brain-Computer Interfaces (BCIs) let people control devices using brain signals, but these devices can be hacked in many new ways. The authors identify five different types of attacks that trick or take over these systems, risking user privacy and safety. They created a tool called EEGle to test and find these kinds of attacks automatically. Their work shows that even people without expert skills can use AI to hack BCIs and that these attacks are hard to detect.
What this means in practice
- •For device security teams: Test BCI devices for vulnerabilities across five attack types using the EEGle framework to enhance user safety.
- •For cybersecurity firms: Develop specialized AI-powered defenses against brain signal hijacking and backdoors in BCIs.$Commercial implications: Enables new security services and products protecting consumer and medical BCI devices from sophisticated AI-driven attacks.
Authors
Zahra Tarkhani, Georgios Akkogiounoglou, Lorena Qendro, Isabel Tscherniak, Anil Madhavapeddy
Abstract
The rapid integration of AI into human-centred systems such as Brain-Computer Interfaces (BCIs) has created a poorly understood attack surface linking neural signals to physical systems. Exploits in this domain threaten cognitive autonomy, mental privacy, and physical safety, from neural data exfiltration to malicious control of BCI-tethered devices. We introduce the NERVE Attacks class, a systematic characterisation of five orthogonal attack dimensions that together span the complete BCI stack: Neuro-mimetic Forgery (N), Evasion via Desynchronization (E), Replay-based Hijacking (R), Vein Tapping (V), and Embedded Backdoors (E). To evaluate this class, we present EEGle, an AI-assisted extensible framework for systematic BCI security analysis. Our evaluation uncovers 17 novel neuro-specific attack instances and reveals a stealth-effectiveness spectrum unique to BCI backdoor design. We also show that generative AI lowers the barrier to entry for non-expert attackers and provide EEGle to the community for building and verifying the security of these deeply personal devices.