5 AI Scams Poised to Surge in 2025—and How to Stay Safe

AI scams are rapidly evolving—and 2025 is shaping up to be a turning point for their growth and sophistication. While deepfakes, voice clones, and AI-powered phishing scams started making headlines in recent years, these were just early warning signs. Fraud experts now believe that as AI tools become more advanced, scammers will escalate their tactics to exploit individuals, businesses, and governments on a massive scale.

According to the Deloitte Center for Financial Services, AI-enabled fraud could result in $40 billion in losses by 2027, a steep rise from $12.3 billion in 2023. Criminals are moving quickly, using AI to boost the believability and scale of their schemes. Platforms like Telegram—a hub for scam coordination—have seen a 644% increase in messages discussing AI-based fraud from last year, signaling just how widespread and organized these efforts have become.

How AI Scams Are Evolving

Here’s what to expect in 2025 and how you can protect yourself:

1. AI-Driven Business Email Compromise (BEC) Scams

Business Email Compromise attacks, already a significant cyber threat, are becoming even more dangerous with AI deepfakes. Scammers are now using AI-generated video and audio to impersonate executives during video calls.

In Hong Kong, fraudsters pulled off two separate schemes where AI impersonations on Zoom convinced employees to transfer a staggering $30 million in funds. This is not an isolated trend—U.S. firm Medius reported that over 53% of accounting professionals were targeted by AI-based deepfake attacks in the past year.

VIPRE Security Group reports that 40% of Business Email Compromise (BEC) emails are now entirely AI-generated. Usman Choudhary, Chief Product and Technology Officer, explained, “As AI technology continues to advance, the potential for BEC attacks is growing exponentially.”

As AI tools continue to improve, BEC scams are expected to grow in both scale and complexity, targeting companies worldwide.

2. Romance Scams Run by AI Chatbots

AI chatbots are changing the face of romance scams. Scammers can now automate conversations with victims, making their schemes appear more convincing than ever. In a shocking example, a Nigerian cybercriminal demonstrated a chatbot that impersonated a military doctor, building trust with a victim who believed she was talking to her love interest.

Unlike traditional scams that relied on human operators, these AI chatbots are:

Fluent, eliminating accents that might raise suspicion.

Available 24/7, enabling scammers to scale operations globally.

Experts predict these fully autonomous chatbots will proliferate in 2025, creating a new wave of AI-powered romance fraud.

3. Pig Butchering Scams Enhanced by AI

“Pig butchering” scams—where victims are groomed over time for large financial fraud—are increasingly powered by AI. Criminal syndicates now use AI tools like “Instagram Automatic Fans” to send thousands of messages a minute, hooking unsuspecting victims with generic introductions like “My friend recommended you. How are you?”

Once trust is established, AI deepfake tools enable scammers to impersonate people on video calls or use voice cloning to make the scam more believable. Fraud experts warn that pig butchering operations will leverage these technologies to scale dramatically in 2025.

Picture a wall of mobile phones, each relentlessly sending thousands of scam messages to victims around the world every minute. Videos circulating on social media and Telegram reveal this staggering reality—a tactic that scam compounds are now using to scale their operations to unprecedented levels.

4. Deepfake Extortion Targeting Executives

High-profile individuals are particularly vulnerable to AI-based extortion. In Singapore, scammers targeted 100 public servants—including cabinet ministers—by creating deepfake videos of them in compromising situations. The scammers demanded $50,000 in cryptocurrency to prevent the videos’ release.

This scheme relied on publicly available images, such as LinkedIn photos and YouTube videos, to generate believable deepfakes. As deepfake tools become more accessible, experts predict that corporate executives worldwide will increasingly become targets of similar extortion scams in 2025.

5. Deepfake Financial Crimes and “Digital Arrest” Scams

In India, over 92,000 cases of deepfake-based financial scams have been reported since January 2024. These scams typically involve criminals impersonating federal law enforcement officials to psychologically manipulate victims into paying fines or ransoms. Using AI-generated videos and voice clones, scammers create convincing “legal emergencies” to pressure victims into transferring large sums of money.

While banks and fintechs race to build defenses, criminals are advancing even faster. With AI tools available for as little as $20 a month and fraud-as-a-service operations growing rapidly, the threat is evolving at an alarming pace. The future of AI scams is here—and it may sound eerily like your own voice.

Cybersecurity experts believe that this trend, already prevalent in Southeast Asia, could spread to Western countries in 2025, creating a significant new fraud risk.

Staying Safe: Practical Steps to Protect Yourself

As AI scams grow more sophisticated, there are still effective ways to defend against them:

• Verify Unexpected Communications: If you receive an urgent email, text, or call asking for personal information or money, hang up and verify the request directly—such as by calling the organization’s official number, NOT the one that is on your phone or in the text or email.

• Challenge Suspicious Calls or Videos: If a family member calls claiming an emergency or claiming they have been kidnapped, ask personal questions only they would know. Use safe words with loved ones to confirm their identity. On video calls, suspicious behavior like glitches or failing to perform simple tasks (e.g., standing up or waving a hand) could indicate a deepfake.

• Be Cautious with Generic Messages on Social Media: Scams often start with vague introductions or unsolicited messages. Avoid engaging unless you’re sure of the sender’s identity.

What This Means for AI and Cybersecurity

The rise of AI scams represents a pivotal moment in the evolution of financial fraud. Criminal organizations are already deploying sophisticated AI tools to enhance their operations, outpacing many existing cybersecurity defenses.

However, awareness and vigilance remain powerful tools. By staying informed about emerging threats—like BEC attacks, romance chatbots, and deepfake extortion—consumers and businesses can better protect themselves. While AI has immense potential for good, its misuse is a reminder of the importance of robust safeguards as the technology continues to advance.

The future of AI-powered scams may be here, but with knowledge and preparation, we can work to stay one step ahead.

Editor’s Note: This article was created by Alicia Shapiro, CMO of AiNews.com, with writing, image, and idea-generation support from ChatGPT, an AI assistant. However, the final perspective and editorial choices are solely Alicia Shapiro’s. Special thanks to ChatGPT for assistance with research and editorial support in crafting this article.