Social engineering Move quickly, at the speed of the generative II. This offers bad actors several new tools and methods of research, survey and operation of organizations. In recent communication, the FBI noted: “As the technology continues to develop,” cybercriminals “are also doing.
This article studies some consequences of this acceleration, which begins from the Genoa. And he studies what that means for IT Liders responsible for the control of protection and mitigating vulnerabilities.
More realism, the best base and scripts of multiple attacks
Traditional social engineering methods usually provide for anyone who knows the goal. The attacker can hide behind the e -mail for communication by adding some psychological triggers to enhance the chances of successful violation. Maybe the request is urgently acting, so the goal is less likely to pause and develop doubts. Or the e -mail compulsion comes from the CEO of an employee, hoping that the employee’s respect means they don’t question the message.
If you use a voice, then the attacker may pretend that the goal is not talking to (and would recognize the voice). Maybe pretending to be from another department or external partner.
Of course, these methods often fall apart if the goal wants to somehow check their identity. Let it want to check their appearance, or how they write in real time.
However, now that the gate has entered the conversation, everything has changed.
Deepfake video growth means that the enemy no longer needs to hide behind the keyboards. They combine valid records for the analysis and reproduction of man’s manner and speech. Then it’s just a deepfake referral to say something or use it as a digital mask that loses what the attacker says and does in front of the camera.
Growth in digital work, with remote workers who were used for virtual meetings means that it is easier to explain possible warning signs. Unnatural movements, or voice sound a little different? Blame it for a bad connection. Speaking to face to face, it adds a layer of authenticity that supports our natural instinct to think that “seeing it”.
The voice cloning technology means that the attackers can also talk in any voice, conducting voice phishing, also known as Wing, attacks. The growth of the ability of this technology is displayed in Open the AI recommendation For banks that have started authentication based on votes as a security measure to access bank accounts and other confidential information. “
The textual communication is also transformed with Genai. LLM growth allows malicious actors to work at almost native level, and the results can be prepared on regional dialects for even greater stagger. This opens the door to new markets for social engineering attacks, when choosing the goals of the language is no longer a blocker.
Adoption of the order on unstructured OSINT from Genai
If anyone was on the Internet, they leave a digital mark somewhere. Depending on what they share, sometimes it can be enough to reveal enough information to bring themselves or endanger their identity. They can share their birthday on Facebook, post their jobs on LinkedIn and put photos of their home, family and life on Instagram.
These actions offer ways to create profiles for use with social engineering attacks on people and organizations they are related. In the past, the collection of all this information would be a long and manual process. Looking for every social media channel, trying to join the points between people’s messages and public information.
Now AI can do it all on Hyperspeed by viewing the Internet for unstructured data, for obtaining, organizing and classification of all possible matches. These include personality recognition systems where you can download a photo of someone and allow the search engine to find all the places they appear on the Internet.
Moreover, as the information is available publicly, you can access and combine this information anonymously. Even when using paid tools for Genai, stolen accounts are sold in a dark network, giving the attackers another way to hide their activities, use and requests.
Turning the data herbs into a treasure herb
Large-scale data leaks are the fact of modern digital life: with more than 533 million users on Facebook, which have details (including birthdays, phone numbers, places), compromised in 2021, up to more than 3 billion Yahoo users who have Sensitive information provided in 2024. Of course, in 2024. Of course.
Instead, people can use Genai tools to autonomously sort large volumes of content. They can find any data that could be used angrily, for example, for extortion, armed weapons of private discussions or theft of intellectual property hidden in the documents.
AI also reflects the creators of the documents (using the form of the specified confession) to establish any outcrying links between different parties, including the transfer of wires and confidential discussions.
Many tools are open source, allowing users to customize plugins and modules. For example, Recon-NG can be customized for use cases such as e-mail and OSINT collection. Other tools are not for public use, such as Red Reaper. It is an AI espionage form capable of sifting hundreds of thousands of letters to discover secret information that can be used against organizations.
Genai Genie comes out of the bottle – is your business being exposed?
Now attackers can use the Internet as a database. They just need a part of the data as a starting point, such as a name, email address or image. Genai can start work by starting real -time requests to go out, reveal and process communication and relationships.
Then it is about choosing the appropriate exploits, often scale and autonomous. Be it video Deepfake and voice cloning, or attacks based on LLM. They would be limited by a selected group of specialists with the necessary knowledge. Now the landscape is democratized with the lifting of “hacking as a service”, which makes most hard work for cybercriminals.
So, how can you know what potentially compromised information is available in your organization?
We’ve built Monitoring threats The tool that tells you. It crawls in every corner of the Internet, reporting what data is there and can be used to create effective attacks, so you can take action before the attacker first got into it.
