You have been warned – this nightmare is now real.
We were notified. Forget the search for Telltale panels, the last set of attacks fueled by the AI is so sophisticated that you must check everything to make sure you are not attacked. In the past 24 hours, we have seen users of Gmail and Outlook have warned that malicious emails are now so “perfect” that they are impossible to detect, and that the calls that seem to come from people we know, could be a dangerous disappointment.
This is the last to come from the FBIAfter the discovery of “a malicious messaging campaign and vocal messaging in progress”. This used texts and voice messages claiming to come from “high American officials”, deceptive victims, many of whom are also “current or former senior officials of the federal government or states and their contacts”.
The office warning is serious enough for you to be said now: “If you receive a pretending message to be a senior American official, do not assume that it is authentic.” The purpose of the attacks is to steal identification information through links that seem to be linked to the message.
According to Max Gannon de Cofense, “It is important to note that threat actors can also usurp known telephone numbers of trust or people, adding an additional layer of attack. The actors of the threat are turning more and more towards AI to carry out phishing attacks, which makes these scams more convincing and almost indistinctable. ”
FBI’s advice is wider than the latter attack, and refers to its recent warnings On the proliferation of AI attacks.
- Before responding to calls or responding to messages, “check the identity of the person who calls you or sends text or voice messages.”
- Check the email addresses, coordinates and URLs for all revealing errors – although AI means that these shifts are rare and replicas can be almost perfect.
- Obviously, beware of all “subtle imperfection in images and videos, such as deformed hands or feet, unrealistic facial characteristics, indistinct or irregular faces, unrealistic accessories such as glasses or jewelry, inaccurate shadows, alerts, the lag of the vocal call, the vocal assortment and the non-natural movements.”
- The same goes for voices. “Listen carefully to the tone and the choice of words to distinguish a legitimate phone call or a vocal message from a known contact and a vocal cloning generated by AI, because they may seem almost identical.”
That said, the FBI recognizes that “the content generated by AI has advanced to the point that it is often difficult to identify”. Sometimes it comes down to common sense. Is this a call that I could reasonably expect, and I asked to do something that would benefit from a cybercriminated or a crook. Can I deduce what they could be taken. How can I hang up and recall with normal channels. How to check the caller.
Ryan Sherstobitoff of SecurityScorecard said to me “to mitigate these risks, individuals must adopt an increased feeling of skepticism towards unlined communications, in particular those which require sensitive information or exhorting immediate action.”
Often these texts, calls and voice messages lead to a link. This is the attack, which will draw information on identification information or encourage you to install malware. “Do not click on any link in an email or SMS before independently confirms the identity of the sender,” warns the office. And “Never open up attachments, click links in messages or download apps on demand or someone you have not checked.”
