ThermoSecure: Investigating the Effectiveness of AI-Driven Thermal Attacks on Commonly Used Computer Keyboards

Abstract

Thermal cameras can reveal heat traces on user interfaces, such as keyboards. This can be exploited maliciously to infer sensitive input, such as passwords. While previous work considered thermal attacks that rely on visual inspection of simple image processing techniques, we show that attackers can perform more effective AI-driven attacks. We demonstrate this by presenting the development of ThermoSecure, and its evaluation in two user studies (N=21, N=16) which reveal novel insights about thermal attacks. We detail the implementation of ThermoSecure and make a dataset of 1,500 thermal images of keyboards with heat traces resulting from input publicly available. Our first study shows that ThermoSecure successfully attacks 6-symbol, 8-symbol, 12-symbol, and 16-symbol passwords with an average accuracy of 92%, 80%, 71%, and 55% respectively, and even higher accuracy when thermal images are taken within 30 seconds. We found that typing behavior significantly impacts vulnerability to thermal attacks, where hunt-and-peck typists are more vulnerable than fast typists (92% vs 83% thermal attack success if performed within 30 seconds). The second study showed that the keycaps material has a statistically significant effect on the effectiveness of thermal attacks: ABS keycaps retain the thermal trace of users presses for a longer period of time, making them more vulnerable to thermal attacks, with a 52% average attack accuracy compared to 14% for keyboards with PBT keycaps. Finally, we discuss how systems can leverage our results to protect from thermal attacks, and present 7 mitigation approaches that are based on our results and previous work.

Conducting and Mitigating Portable Thermal Imaging Attacks on User Authentication using AI-driven Methods

“…automatically detect keyboards and key-presses in a thermal image [7,8]. This demonstration will display our current research, utilising the same automatic detection to create thermal cameras systems which obfuscate authentication interfaces, preventing thermal attacks as they happen and whether or not they were detected by the potential victim.…”

“…First investigated in 2011 by Mowery [16], research in the years since have repeatedly found thermal imaging effective at conducting side-channel attacks on keyboards [8], keypads [14] and touchscreens [1]. Unlike shoulder surfing attacks which require real-time observation of user authentication, thermal attacks can be conducted in the window of time after the victim has entered their details and then left their device unattended, making them easier to perform.…”

“…Consumers can purchase a thermal camera for only $150 to discretely augment their smartphones, and researchers have successfully integrated thermal cameras into wearable glasses and displays [2,3,10] (see Figure 1). This raises security concerns, however, as thermal cameras have been shown capable of enabling thermal attacks [1,8,9,16]. This category of effective side-channel attacks functions by capturing a thermal image of a keyboard, keypad or touchscreen following user authentication, which then reveals heat traces that indicate the keys or areas pressed, and in what order, allowing deduction of the user's password or pattern.…”

PIN Scrambler: Assessing the Impact of Randomized Layouts on the Usability and Security of PINs

“…Threat modeling is often defined as the formal process of identifying, documenting, and mitigating security threats to a system [20]. Human-centered threats involve social engineering attacks [15], shoulder surfing attacks [9,11], thermal attacks [1,2,5], smudge attacks [7,24], just to name a few. Designing authentication schemes against multiple threats is challenging, mainly because protecting users against all security and privacy threats is close to impossible [18].…”

Vulnerability Analysis and Security Assessment of Secure Keyboard Software to Prevent PS/2 Interface Keyboard Sniffing

“…Ref. [ 16 ] introduced AI-driven thermal attacks on commonly used computer keyboards. However, this attack technique has a limitation in that it requires thermal cameras.…”