Incident: Smart Speakers Vulnerable to Laser Attacks for Unauthorized Control.

Published Date: 2019-11-05

Postmortem Analysis
Timeline 1. The software failure incident of smart speakers being vulnerable to lasers was reported on November 5, 2019 [Article 92135]. 2. Published on 2019-11-05 08:00:00+00:00. 3. The incident occurred on November 5, 2019.
System 1. Amazon's Alexa 2. Google Assistant 3. Apple's Siri 4. Google Home 5. iPhone XR 6. Samsung Galaxy S9 7. Google Pixel 2 [92135]
Responsible Organization 1. The researchers based in Tokyo and at the University of Michigan were responsible for causing the software failure incident by discovering the vulnerability of smart speakers to lasers [92135].
Impacted Organization 1. Amazon 2. Google 3. Apple 4. Ford 5. Tesla [Cited from Article 92135]
Software Causes 1. Vulnerability in smart speakers' microphones reacting to light aimed directly at them, causing them to make electrical signals as if they're hearing someone's voice [92135].
Non-software Causes 1. The susceptibility of smart speakers to lasers, which allowed researchers to take over devices' microphones by hitting them with beams of light [92135].
Impacts 1. The vulnerability discovered by researchers allowed for the takeover of smart speakers such as Amazon's Alexa, Google Assistant, and Apple's Siri by using lasers to manipulate the microphones, potentially enabling hackers to perform actions like opening garage doors remotely and controlling smart home devices [92135]. 2. The vulnerability also extended to smartphones like the iPhone XR, Samsung Galaxy S9, and Google Pixel 2, although the proximity required for control was closer compared to smart speakers [92135]. 3. The potential impacts of this software failure incident include unauthorized online purchases, control over smart home switches, and the ability to remotely unlock and start cars linked to the compromised devices [92135].
Preventions 1. Implementing additional authentication measures for sensitive actions such as opening garage doors or making online purchases could have prevented the incident [92135]. 2. Conducting thorough security testing, including testing for vulnerabilities related to unconventional attack methods like using lasers on microphones, could have helped identify and address this vulnerability before it could be exploited [92135]. 3. Regularly updating the software and firmware of smart devices to patch any known security vulnerabilities could have mitigated the risk of such attacks [92135].
Fixes 1. Implementing additional security measures to detect and prevent unauthorized access through light commands, such as enhancing microphone sensitivity settings or incorporating light-blocking materials [92135].
References 1. Researchers based in Tokyo and at the University of Michigan 2. The New York Times 3. Google spokesperson 4. Amazon spokesperson 5. Ford and Tesla (did not respond to requests for comment) [92135]

Software Taxonomy of Faults

Category Option Rationale
Recurring multiple_organization (a) The software failure incident related to the vulnerability of smart speakers to lasers has happened at multiple organizations. The incident affected Amazon's Alexa, Google Assistant, and Apple's Siri [92135]. The researchers also tested this method on devices like the iPhone XR, Samsung Galaxy S9, and Google Pixel 2 [92135]. The researchers informed Amazon, Apple, Google, Ford, and Tesla about this potential security issue [92135]. Google and Amazon have responded by stating that they are closely reviewing the research and engaging with the researchers to understand their work [92135]. Apple, Ford, and Tesla did not offer comments or respond to requests for comment [92135].
Phase (Design/Operation) design (a) The software failure incident in this case is related to the design phase. Researchers discovered a vulnerability in smart speakers such as Amazon's Alexa, Google Assistant, and Apple's Siri, where they could be taken over by hitting the devices' microphones with beams of light. This vulnerability was due to the fact that microphones also react to light aimed directly at them, in addition to sound. The incident was a result of a design flaw in the smart speakers that allowed the microphones to be tricked into making electrical signals as if they were hearing someone's voice [92135]. (b) The software failure incident is not related to the operation phase or misuse of the system.
Boundary (Internal/External) within_system (a) within_system: The software failure incident described in the article is primarily within the system. The vulnerability exploited by the researchers involved tricking the microphones of smart speakers and other devices into reacting to light aimed directly at them, causing them to make electrical signals as if they were hearing someone's voice. This manipulation of the microphones' functionality is an internal system vulnerability that the researchers were able to exploit [92135].
Nature (Human/Non-human) non-human_actions (a) The software failure incident in this case is related to non-human actions. Researchers discovered a vulnerability in smart speakers where they could be taken over by hitting the devices' microphones with beams of light, tricking the microphones into making electrical signals as if they're hearing someone's voice. This vulnerability was not introduced by human actions but rather by the susceptibility of the devices to lasers [92135].
Dimension (Hardware/Software) hardware, software (a) The software failure incident in this case is related to hardware. Researchers discovered a vulnerability in smart speakers such as Amazon's Alexa, Google Assistant, and Apple's Siri, where they could be taken over by hitting the devices' microphones with beams of light [92135]. The vulnerability exploited the fact that microphones also react to light aimed directly at them, tricking them into making electrical signals as if they're hearing someone's voice [92135]. (b) The software failure incident is also related to software. The vulnerability allowed hackers to potentially use light commands to buy stuff online, control smart home switches, and remotely unlock and start a car linked to the speaker [92135]. The incident highlights a software flaw in the way these smart devices process input signals from their microphones, allowing unauthorized access and control.
Objective (Malicious/Non-malicious) malicious (a) The software failure incident described in the article is malicious in nature. Researchers discovered a vulnerability in smart speakers that could be exploited by hitting the devices' microphones with beams of light, allowing hackers to take over Amazon's Alexa, Google Assistant, and Apple's Siri. The hackers could potentially use this method to buy items online, control smart home switches, and even remotely unlock and start a car linked to the speaker [92135].
Intent (Poor/Accidental Decisions) unknown The intent of the software failure incident related to the vulnerability discovered in smart speakers being susceptible to lasers was not due to poor decisions or accidental decisions. Instead, it was a result of a novel discovery by researchers who found that microphones in smart speakers react to light aimed directly at them, in addition to sound. The incident was not caused by poor decisions or accidental decisions but rather by a previously unknown vulnerability in the technology [92135].
Capability (Incompetence/Accidental) accidental (a) The software failure incident in this case is not directly related to development incompetence. The vulnerability discovered by the researchers was unexpected and involved exploiting a physical characteristic of the devices (microphones reacting to light) rather than a failure in the development process [92135]. (b) The software failure incident can be categorized as accidental. The vulnerability that allowed the smart speakers to be taken over by lasers was not intentionally designed into the devices but was an accidental discovery by the researchers. It was not a deliberate feature or flaw in the software but rather an unforeseen susceptibility to a specific external stimulus (light) [92135].
Duration temporary (a) The software failure incident described in the article is more likely to be temporary rather than permanent. The vulnerability discovered by the researchers involved using lasers to manipulate smart speakers and devices like the iPhone XR, Samsung Galaxy S9, and Google Pixel 2. The vulnerability was specific to the method of using light to trick the microphones into making electrical signals, allowing hackers to potentially control various functions of the devices. The incident was not a result of inherent design flaws in the devices but rather a specific vulnerability that could be addressed with appropriate security measures. The companies involved, such as Amazon, Apple, Google, Ford, and Tesla, were informed about the issue, indicating that steps could be taken to mitigate the vulnerability [92135].
Behaviour omission, value, other (a) crash: The articles do not mention any instances of the software crashing. (b) omission: The vulnerability discovered by researchers involved the smart speakers' susceptibility to lasers, which could allow hackers to take over devices like Amazon's Alexa, Google Assistant, and Apple's Siri by tricking the microphones into making electrical signals as if they're hearing someone's voice. This could lead to the omission of the system's intended functions as the hackers could remotely control various actions like buying stuff online, controlling smart home switches, and unlocking and starting a car linked to the speaker [92135]. (c) timing: The articles do not mention any instances of the software performing its intended functions too late or too early. (d) value: The vulnerability discovered by researchers could lead to the system performing its intended functions incorrectly, as hackers could manipulate the microphones of smart speakers to carry out unauthorized actions like opening a garage door or controlling smart home devices [92135]. (e) byzantine: The articles do not mention any instances of the system behaving erroneously with inconsistent responses and interactions. (f) other: The other behavior observed in this software failure incident is the system's susceptibility to external manipulation through a novel method involving lasers, which was not a typical scenario considered for security threats [92135].

IoT System Layer

Layer Option Rationale
Perception sensor (a) The failure was related to the perception layer of the cyber physical system that failed due to contributing factors introduced by sensor error. The vulnerability exploited by the researchers involved tricking the microphones of smart speakers and devices like the iPhone XR, Samsung Galaxy S9, and Google Pixel 2 into reacting to light aimed directly at them, making electrical signals as if they're hearing someone's voice. This manipulation allowed the researchers to take control of the devices, demonstrating a sensor-related failure [92135].
Communication link_level The software failure incident reported in the article [92135] is related to the link_level of the cyber physical system. The vulnerability discovered by the researchers involved using lasers to manipulate the microphones of smart speakers and devices like the iPhone XR, Samsung Galaxy S9, and Google Pixel 2. By focusing lasers on the microphones, the researchers were able to trick the devices into responding as if they were receiving voice commands, enabling them to perform actions such as opening a garage door remotely or controlling smart home switches. This manipulation of the physical layer (microphones reacting to light) highlights a vulnerability in the communication link between the physical and digital components of these devices.
Application TRUE The software failure incident described in the article [92135] is related to the application layer of the cyber physical system. The vulnerability exploited by the researchers involved tricking the microphones of smart speakers and other devices into reacting to light aimed directly at them, which caused the devices to respond as if they were receiving voice commands. This manipulation allowed the researchers to take control of various functions such as opening a garage door, buying items online, controlling smart home switches, and remotely unlocking and starting a car linked to the speaker. This exploitation of the microphones' response to light represents a failure at the application layer, as it involves manipulating the devices' functionality through an unexpected method introduced by the researchers [92135].

Other Details

Category Option Rationale
Consequence property, non-human, theoretical_consequence (d) property: People's material goods, money, or data was impacted due to the software failure The software failure incident described in the articles involved a vulnerability in smart speakers that allowed hackers to take control of devices like Amazon's Alexa, Google Assistant, and Apple's Siri by using lasers to manipulate the microphones. This vulnerability could potentially allow hackers to make unauthorized purchases online, control smart home switches, and even remotely unlock and start a car linked to the speaker. The researchers also tested this method on smartphones like the iPhone XR, Samsung Galaxy S9, and Google Pixel 2, although they had to be much closer to the devices to take control of them [92135].
Domain information (a) The software failure incident reported in the article is related to the industry of information. The vulnerability discovered by researchers allowed them to take over smart speakers like Amazon's Alexa, Google Assistant, and Apple's Siri by using lasers to manipulate the microphones, enabling them to control various functions such as buying online, controlling smart home switches, and even remotely unlocking and starting a car linked to the speaker [Article 92135].

Sources

Back to List