Security Alert Issued As Hackers May Resort To Machine Learning To Exploit New Text Messaging Vulnerability

While machine learning models are the rage these days, thanks to the growing trend of AI, the increasing number of threats on this front cannot be underestimated.

Today, we’re seeing security researchers announce a new threat that was recently discovered and is related to text messaging. Moreover, this may force hackers to track down a user’s location.

The news comes thanks to Ph.D. researchers from Northeastern who brought to light a new flaw by using machine learning technology to attain data from a primitive SMS program. The latter has paved the way for texts in smartphones since the start of the 1990s. Moreover, the work may be located on certain servers as well such as arXiv.

The experts revealed that all you need are phone number belonging to the victim before you can get the right type of access to locate them. While the research report is yet to be reported at a top Security Symposium in California, we know that there are some great findings that are useful for obvious reasons.

The end result is greater tracking of users linked to various locations around the globe.

With time, SMS security continues to improve, little by little. And that’s since it was originally begun for the likes of 2g systems, around 30 years back.

Therefore, whenever a message is sent out to the user, the phone would generate a response with the help of alerts produced to senders, including a delivery receipt.

Through such means, hackers would produce a long list of messages to the smartphone as well. And then the timing for replies of the delivery would allow threat actors to locate the exact location of the victim, no matter if the communications get encrypted or not.

Such timings of the alerts for automated delivery being sent out by a device would end up leaving hints of the location via fingerprints. And that was fine until you realize that machine learning models use such means to create algorithms that can detect it.

After being created, hackers can release several texts. The results are directly put into models that would respond using estimated locations.

For now, the experts who deal with cybersecurity are yet to delineate any links of the vulnerability, that’s seen in Android systems, to be exploited right now.

But if you think that news is comforting, then think again. Hackers can easily make use of this behavior at a later point in time and by then, it’s going to be hard to track down how exactly it arose.

Attackers will require the use of several Android smartphones in different settings and be ready to produce messages each hour by calculating such a response. And in terms of the collection process, it would take some time, all related to the figure of fingerprints which attackers need to collect.

Experts claim that those people who prefer to keep locations private, including activists, leaders of firms, and top government officials, are more likely to be impacted by this ordeal.

Read next: The Alarming State of Email Security Worldwide
Previous Post Next Post