A trio of vulnerabilities recently found in Alexa devices that could have led to broader attacks on home and corporate networks have been patched in a software update by Amazon.
The vulnerabilities, which were made public last week by researchers at Check Point, raised red flags given the many millions of people now working from home because of the pandemic.
More than 200 million Alexa smart-home devices have shipped to date, according to multiple sources. The vulnerabilities, which could be exploited by clicking on a bad link filled with malicious code, could have exposed personal user info, including banking data histories, usernames, phone numbers, and home addresses.
Oded Vanunu, head of Check Point's product vulnerability research, says security pros should be concerned about the Alexa vulnerabilities because most home users don't have basic network security, such as network segmentation, in place.
"An Alexa account can serve as a gateway to the home network and be used as a base to launch other attacks," Vanunu says. "While segmenting the network is a good idea, it's a lot to ask of the typical home user."
Brandon Hoffman, CISO at Netenrich, agrees that placing the onus on consumers to do more extensive management of the network doesn't make sense.
"Rather, device manufacturers have to recognize that the work-from-home shift puts a greater focus on home equipment, and they have a responsibility to the consumer base to take proper measures in closing vulnerabilities and providing security guidance along with more robust security options for savvy users," Hoffman says.
Hank Schless, senior manager of security solutions at Lookout, says security teams need to understand that traditional tools do not protect their employees from this type of attack. Even if security pros deploy a VPN, multifactor authentication, and mobile device management, none of that will stop an employee from tapping on a convincing phishing link and giving up their corporate login credentials or introducing malware into the corporate infrastructure.
"In the case of a consumer app, app developers need to integrate security into their app to protect their customers from malware," Schless says. "App developers and security teams need to work together to integrate security measures into their mobile apps that protects the user."
Check Point presented the vulnerabilities to Amazon in June. The company was receptive to the research and patched the vulnerabilities right away, Vanunu says. In recent months, Check Point has also conducted security research on TikTok, WhatsApp, and Fortnite.
"Alexa has concerned us for a while now, given its ubiquity and connection to IoT devices," Vanunu says. "We hope manufacturers of similar devices will follow Amazon's example and check their products for vulnerabilities that could compromise users' privacy."
Check Point's researchers found that certain Amazon/Alexa subdomains were vulnerable to Cross-Origin Resource Sharing (CORS) misconfigurations and Cross Site Scripting (XSS). Using XSS, the researchers were also able to get the cross-site request forgery (CSRF) token and perform actions on the victim's behalf.
Vanunu said a single click by a victim triggers the three vulnerabilities: First, the XSS vulnerability in one of Amazon's subdomains enables access to the victims' identification cookies. Second, once access is gained, both the CORS misconfiguration and the CSRF token can be exploited, and, third, actions can be performed on behalf of victims on their Alexa accounts.
A CSRF token is a unique, secret, unpredictable value that's generated by the server-side application and transmitted to the client in such a way that it's included in a subsequent HTTP request made by the client. Vanunu says these exploits could let an attacker remove or install Alexa skills (apps) on a targeted victim's Alexa account, access the person's voice history, and acquire personal information.
While Vanunu reiterated that prevention was mostly the burden of the manufacturer, users can take certain steps to protect their Alexa accounts. For starters, he says, users should not install unfamiliar apps to their systems. They should also think twice before sharing password or bank account information.
"People should also delete their voice histories and know how many apps they've installed," Vanunu added.
Netenrich's Hoffman says while hackers have been prying into home automation equipment for a long time, the real value of penetrating into home networks initially had limited value to cybercriminals. However, the paradigm has changed now that more people work from home.
"The most dangerous consideration of the work-from-home shift is the idea that vulnerable home networks create a bridge into highly valuable corporate networks," Hoffman says. "While the recent example of Alexa-based vulnerabilities is interesting in terms of highlighting the capability, the actual data accessed is of low or very little value to cybercriminals unless obtained on a massive scale. On the other hand, moving laterally or escalating access to machines on the network shared with a device like Alexa has massive appeal."
Chris Morales, head of security analytics at Vectra, says security teams need to finally understand that IoT attacks are real and here for the long-term. He says large-scale DDoS attacks – the original use of IoT botnets – are difficult to combat for even the largest, most prepared businesses.
"An even greater danger is when IoT devices start snooping around corporate networks and can pivot to more critical systems," he says. "Devices, such as virtual assistants, printers, cameras and even advanced devices like MRI scanners, can pose an alarming cybersecurity risk. While they don't fit the bill of a traditional network host, they represent fruitful targets and vectors for cyberattackers."