DeepSec 2024 Talk: AI’s New Era: Impacts on Health Data Security and Beyond – Sina Yazdanmehr & Lucian Ciobotaru
It has become easier to create AI systems because of the availability of many options and datasets. These AIs can quickly gain expert knowledge in different domains, enabling attackers to exploit scientific knowledge and target system and data security, which was not workable before. Although recent studies have highlighted these impacts, a tangible example has been missing. For instance, attackers can use AI’s expert knowledge in the healthcare sector to perform complex attacks with no need of domain expertise.
Earlier this year, Google launched Health Connect, an Android app designed to share data seamlessly between medical and fitness apps, intended to replace Google Fit. While Health Connect is robust against conventional cyberattacks, it is susceptible to these emerging threats.
In this talk, we will show an example of these threats by explaining a malicious app we developed. The app gathers data from Health Connect and sends it to a medical AI, which then crafts fake data tailored to the victims’ medical conditions. This allows us to steer other apps’ output into suggesting incorrect treatments and recommendations without the user noticing. We’ll show how such manipulation could alter diet control, family planning, and diabetes management apps, leading to serious medical issues for the victims.
We’ll conclude with mitigation strategies for developers and technology companies on building AI-resistant technologies and apps.
We asked Sina and Lucian a few more questions about their talk.
Please tell us the top facts about your talk.
- AI is rapidly emerging, transforming various industries, and becoming accessible to everyone
- Its impact on cybersecurity goes beyond enhancing existing attacks
- AI can quickly gain expert knowledge in different domains, empowering attackers to exploit additional aspects of a system.
- In this talk, we will show this threat using Google Health Connect as an example
How did you come up with it? Was there something like an initial spark that set your mind on creating this talk?
Healthcare and digital health security are key areas of focus for us. We’ve conducted extensive research in this field and are always exploring new topics. When we learned about Google Health Connect and its ability to share health data between apps, we saw a potential vulnerability. Given Google’s strong security measures against typical attacks, we began investigating alternative ways to manipulate the data. This led us to explore AI and leverage scientific knowledge for innovative exploitation strategies.
Why do you think this is an important topic?
While many researchers have explored AI’s impact on cybersecurity, some experts and decision-makers still view AI as merely a tool to enhance traditional attacks. What’s often overlooked is AI’s potential to enable entirely new types of attacks, exploiting different facets of a system. This is a crucial fact, and we must keep it in mind when developing AI-resistant systems.
Is there something you want everybody to know – some good advice for our readers, maybe?
AI’s impact on cybersecurity extends far beyond crafting better phishing emails or speeding up the exploitation of common vulnerabilities. Freely available AIs, coupled with extensive datasets, can quickly gain domain-specific knowledge, allowing attackers to exploit that knowledge without moral constraints. This poses serious risks across sectors like healthcare, logistics, and aviation. In this talk, we will show an example of such a threat.
A prediction for the future – what do you think will be the next innovations or future downfalls with your field of expertise / the topic of your talk in particular?
We will see more AI-powered attacks and threats in the future, and we hope to have sufficient countermeasures in place.
Sina Yazdanmehr is a senior information security consultant and researcher. Since 2009, he has worked for different security firms and CERT, developing a strong expertise in cloud, application, and telecom security. He has presented his research at conferences like Black Hat. Recently, his expertise extended to healthcare cybersecurity, discovering structural issues that will be presented at this conference.
Lucian Ciobotaru is a cybersecurity expert with a background in healthcare. After transitioning from medical school to cybersecurity, he developed deep expertise in identifying and addressing security issues in the healthcare sector. Lucian’s recent work focuses on leveraging his medical knowledge to enhance the security of healthcare and digital health systems.