Decoding digital deception: UTA researcher develops framework to combat AI-powered fraud

The digital age has redefined deception. What once arrived as clumsy phishing emails or awkward scam calls is now powered by artificial intelligence, capable of mimicking human voices, generating realistic faces, and crafting flawless written communication.

Dr. Gabriel Aguilar, assistant professor of technical writing and professional communication at The University of Texas at Arlington, is studying how this new generation of fraud leverages AI to outsmart traditional security awareness and how education can serve as a frontline defense.

Aguilar’s recently published study in the Journal of Business and Technical Communication explores the evolution of scams from crude deception to AI-driven manipulation, where technologies such as deepfakes, voice-cloning, and AI chatbots blur the line between real and artificial. His research emphasizes how these technologies, originally designed for productivity and creativity, are now being weaponized to exploit human trust. “AI didn’t revolutionize scamming, it just gave scammers more tools to make their tactics more convincing,” Aguilar said. “It’s not about new motives, but new means.”

According to Aguilar’s research, modern scam networks increasingly depend on generative AI systems that can replicate human communication patterns with near perfection. Using natural language processing (NLP) models, scammers can generate personalized, grammatically flawless messages in any language or dialect, significantly increasing believability.

Deepfake and voice-cloning technologies further enhance deception. By synthesizing voice samples, scammers can now impersonate real people such as employers, relatives, or company executives in live phone conversations. AI-generated videos, meanwhile, can simulate genuine job interviews or official meetings, adding a dangerous visual authenticity.

In contrast, earlier scams like those Aguilar once encountered as a student were easy to flag. Typographical errors, inconsistent formatting, or broken English were clear red flags. The difference now lies in machine precision: today’s AI tools eliminate these telltale signs, enabling scams that are nearly indistinguishable from legitimate digital communication. “If my old scam had been powered by AI, it might have succeeded,” Aguilar reflected. “The check would’ve looked cleaner, the message flawless, and the request far more believable.”

Aguilar’s research also highlights the social side of AI-driven deception. Scammers often target linguistically diverse and economically vulnerable communities, including Latino populations, using culturally familiar language and imagery. AI translation and localization tools make it easier for fraudsters to tailor their approach to specific demographic groups turning diversity into a point of exploitation.

“The precision of AI allows scammers to adapt their tone, language, and even accent,” Aguilar explained. “It's a hyper-personalized deception. And when that’s combined with economic pressure, it becomes a powerful manipulative tool.”

To counter this growing threat, Aguilar has developed a four-part framework that helps educators integrate AI literacy into technical and professional communication courses. His approach trains students to analyze the structure, language, and tone of digital messages skills that help identify subtle inconsistencies or signs of synthetic generation.

Through technical writing instruction, students learn to detect patterns in communication that machines often overlook: emotional cues, logical inconsistencies, or unnatural phrasing. Aguilar’s framework also encourages peer education, where students share their knowledge with families and communities to create a ripple effect of awareness. “When students learn to identify when something feels off, they become multipliers of knowledge,” Aguilar said. “They can help protect their networks against AI deception.”

Aguilar emphasizes that education and vigilance are key to prevention. Recognizing signs of AI synthesis such as unusual voice modulation, mismatched facial movement in videos, or overly formal writing can help users stay alert. Beyond awareness, he calls for collaboration between educators, cybersecurity professionals, and policymakers to create public resources on AI scam detection.

His research suggests combining critical thinking, digital literacy, and writing precision as a defense strategy teaching individuals how to evaluate not just what they read or hear, but how it’s produced.

While AI is often viewed as a neutral tool, Aguilar warns that its creative partnerships and the collaboration between human intent and machine capability can turn dangerous when used maliciously. “In my field, we usually talk about human-AI partnerships producing positive outcomes,” he said. “But scammers are forming the same partnerships just for unethical purposes.”

Aguilar’s work repositions technical writing as more than a career skill. It becomes a cognitive shield, a way of thinking critically about digital interactions in an era when truth can be algorithmically forged. “Technical writing helps dissolve the blinders people wear when faced with something that looks official,” Aguilar said. “It trains them to question, verify, and recognize when technology is being used against them.”

As artificial intelligence continues to evolve, so too do the scams it enables. Aguilar’s research underscores that technology is not just a tool of convenience it is also a medium of persuasion. In this new digital landscape, awareness is the best encryption, and education is the most powerful firewall.

Through AI literacy and critical communication training, Aguilar and his students are redefining cybersecurity not just as a technical challenge, but as a human one.