Eye-tracking systems in augmentative and alternative communication (AAC)
Augmentative and Alternative Communication (AAC) systems are essential for individuals with severe speech or language impairments, including children with autism, people with cerebral palsy, stroke survivors, individuals with ALS, and persons with developmental or intellectual disabilities.
For many users, eye-tracking technology is their primary or only means of communication.
AI enhances these systems by improving accuracy, responsiveness, and usability, enabling more natural and reliable interaction. Given their critical role in daily life, ensuring the security and integrity of these systems is essential to protect users, caregivers, and sensitive personal data.
Security challenges
As a high-risk AI-enabled system operating in a sensitive healthcare context, this technology faces multiple security challenges:
- Protection of sensitive personal and behavioural data
- Integrity of AI models and calibration algorithms
- Hardware-level vulnerabilities in dedicated AI components
- Software supply chain risks
- Real-time reliability and resilience against attacks
Any compromise could disrupt communication, violate privacy, or undermine user trust.
SHASAI’s Contribution
SHASAI strengthens this assistive technology by:
- Applying secure-by-design principles across hardware and software
- Assessing vulnerabilities in AI models and dependencies
- Detecting potential data poisoning or model manipulation
- Enhancing supply chain security
- Enabling automated testing against AI-specific threats
- Supporting continuous monitoring and anomaly detection
Through this validation scenario, SHASAI demonstrates how secure AI methods can protect critical assistive technologies while maintaining performance, safety, and regulatory compliance.