Here’s the thing, social engineering isn’t hackers in hoodies anymore. It’s your daily inbox, your meeting invites, even your phone calls. The cost? The average business lost over $18,000 per successful attack in 2025 alone, and over half these breaches started with a simple trust exploit. Are you ready to spot the new tricks before they land in your feed?
Cybersecurity: The New Social Engineering Threats
Attackers aren’t brute forcing, they’re blending in. The numbers are staggering. Up to 90% of all cyber incidents involve some form of social engineering. But the newest threat is how hackers are using AI as a weapon. They are using AI to generate flawless, personalized messages that mimic colleagues, executives, and trusted brands. These attacks feel timely, urgent, and legit, often personalized using public information from sources like LinkedIn.
4 Dangerous Trends Every Leader Should Spot
- Impersonation via Trusted Brands: Cybercriminals are creating new accounts from legitimate services like QuickBooks or Microsoft 365. They send emails from a real domain that asks for reviews of corrupted databases or urgent invoice issues. Since the sender “looks” legitimate, these messages often bypass technical filters.
- AI Voice and Deepfake Spoofing: Generative AI can clone an executive’s voice or create deepfake video, hitting IT help desks or finance with convincing calls. This type of social engineering attack can cause immense financial and reputational damage. In a recent, well documented scam, a finance analyst at a major firm wired $25 million after receiving what appeared to be a video call with cloned voices of their leadership team.
- Social Media Reconnaissance: Attackers mine LinkedIn profiles, recent hires, and breached data to create highly personalized email and phone scams. By referencing knowledge of a company’s quarterly results or new client wins, they quickly build trust and trick even the savviest employees into responding.
- At Scale Automated Attacks: AI powered bots launch mass phishing campaigns, complete with automated fake system popups and fake browser prompts. They can deliver malware to dozens of devices fast and are designed to exploit confusion and a sense of urgency across an organization.
Real-World Lessons
Most firms spend all year tightening filters but miss that up to 35% of social engineering attacks use legitimate accounts and brands, completely bypassing technical layers. A robust defense starts with team awareness and drills using realistic lures.
Proof Points:
- Over half of social engineering breaches in 2025 led to direct data exposure or business disruption.
- Security teams that ran scenario based drills saw a 34% reduction in employee mistakesthose numbers stick.
- Voice deepfake scams in finance and manufacturing have averaged $1.8 million in loss per incident.
- SubdoMailing is still happening. This is a tactic where attackers use inactive subdomains of trusted brands to send malicious emails.
The Conclusion
These attacks don’t just hit the inbox, they hit the workflow, and the cost rises faster than most leaders expect. The one step every business needs? Start with company wide awareness, practical scenario training, and help desk policies that demand verification, not just speed. Ready for next-level defense? Book a 15-minute strategy call, let’s build a proactive playbook.
FAQ
- Q: How can staff spot a voice deepfake? Ask for a callback on a known phone number, and never trust urgency. If a request feels “off,” it probably is. Pay attention to subtle audio glitches, a monotone voice, or strange phrasing.
- Q: What’s the most overlooked threat in social engineering now? Phishing based on publicly available information from social media and professional networks. Many executives are targeted via personal messages tied to their real-world professional contacts.
- Q: Why aren’t filters enough? At least 35% of social engineering attacks use legitimate accounts and real brands, which can easily slip past standard email and spam filters. A human must be the last line of defense.
- Q: How does AI change cybersecurity? AI is used as a weapon by hackers to create hyper realistic emails, voice clones, and automated scams, making their attacks far more convincing and effective than ever before. It’s the biggest new challenge facing cybersecurity today.
- Q: What is a subdomain takeover, and how can it be used to exploit my brand’s trust? Cybercriminals use inactive subdomains to exploit the trust of legitimate URL. Many companies fail to remove DNS records for inactive domains, and take over these domains and send malicious content that looks completely trustworthy.
Sources
- https://unit42.paloaltonetworks.com/2025-unit-42-global-incident-response-report-social-engineering-edition/
- https://onsecurity.io/article/linkedin-targeted-by-phishing-scams/
- https://www.rapid7.com/blog/post/3-ways-social-engineering-is-evolving-and-what-security-teams-must-do-next/
- https://www.gasa.org/post/guardio-labs-exposes-enormous-subdomailing-scheme-exploiting-trusted-brands-in-a-digital-fraud-cri
