Social engineering is one of the most common attack vectors used by cyber criminals to enter companies. These manipulative attacks often occur in four stages: 

1. Info stealing from targets
2. Building relationships with target and earning trust
3. Exploitation: Convincing the target to take an action
4. Execution: Collected info is used to launch attack 

Five Intelligence Sources

So, how do attackers collect information about their targets? Cybercriminals can employ five types of intelligence to obtain and analyze information about their targets. They are:

1. OSINT (open-source intelligence)

OSINT is a hacking technique used to gather and evaluate publicly available information about organizations and their employees. 

OSINT technologies can help threat actors learn about their target's IT and security infrastructure, exploitable assets including open ports and email addresses, IP addresses, vulnerabilities in websites, servers, and IoT (Internet of Things) devices, leaked or stolen passwords, and more. Attackers use this information to conduct social engineering assaults.

2. Social media intelligence (SOCMINT)

Although SOCMINT is a subset of OSINT, it is worth mentioning. Most people freely provide personal and professional information about themselves on major social networking sites, including their headshot, interests and hobbies, family, friends, and connections, where they live and work, current job positions, and a variety of other characteristics. 

Attackers can use SOCINT software like Social Analyzer, Whatsmyname, and NameCheckup.com to filter social media activity and information about individuals to create tailored social engineering frauds. 

3. ADINT (Advertising Intelligence)

Assume you download a free chess app for your phone. A tiny section of the app displays location-based adverts from sponsors and event organizers, informing users about local players, events, and chess meetups. 

When this ad is displayed, the app sends certain information about the user to the advertising exchange service, such as IP addresses, the operating system in use (iOS or Android), the name of the mobile phone carrier, the user's screen resolution, GPS coordinates, etc. 

Ad exchanges typically keep and process this information to serve appropriate adverts depending on user interests, behavior, and geography. Ad exchanges also sell this vital information. 

4. DARKINT (Dark Web Intelligence)

The Dark Web is a billion-dollar illegal marketplace that trades corporate espionage services, DIY ransomware kits, drugs and weapons, human trafficking, and so on. The Dark Web sells billions of stolen records, including personally identifiable information, healthcare records, financial and transaction data, corporate data, and compromised credentials. 

Threat actors can buy off-the-shelf data and use it for social engineering campaigns. They can even hire professionals to socially engineer people on their behalf or identify hidden vulnerabilities in target businesses. In addition, there are hidden internet forums and instant messaging services (such as Telegram) where people can learn more about possible targets. 

5. AI-INT (artificial intelligence)

In addition to the five basic disciplines, some analysts refer to AI as the sixth intelligence discipline. With recent breakthroughs in generative AI technologies, such as Google Gemini and ChatGPT, it's easy to envisage fraudsters using AI tools to collect, ingest, process, and filter information about their targets. 

Threat researchers have already reported the appearance of dangerous AI-based tools on Dark Web forums such as FraudGPT and WormGPT. Such technologies can greatly reduce social engineers' research time while also providing actionable information to help them carry out social engineering projects. 

What Can Businesses Do to Prevent Social Engineering Attacks?

All social engineering assaults are rooted in information and its negligent treatment. Businesses and employees who can limit their information exposure will significantly lessen their vulnerability to social engineering attacks. Here's how.

Monthly training: Use phishing simulators and classroom training to teach employees not to disclose sensitive or personal information about themselves, their families, coworkers, or the organization.

Draft AI-use policies: Make it plain to employees what constitutes acceptable and unacceptable online activity. For example, it is unacceptable to prompt ChatGPT with a line of code or private data, as well as to respond to strange or questionable queries without sufficient verification.

Utilize the same tools that hackers use: Use the same intelligence sources mentioned above to proactively determine how much information about your firm, its people, and its infrastructure is available online. Create a continuous procedure to decrease this exposure.

Good cybersecurity hygiene begins with addressing the fundamental issues. Social engineering and poor decision-making are to blame for 80% to 90% of all cyberattacks. Organizations must prioritize two objectives: limiting information exposure and managing human behavior through training exercises and education. Organizations can dramatically lower their threat exposure and its possible downstream impact by focusing on these two areas.