Part 7/9:
Researchers and security experts have tested these infiltration methods firsthand by creating a "honeypot" job listing, intentionally open to many applicants. Through this controlled experiment, they observed operatives using AI assistance and deepfakes to appear legitimate.
In one recorded simulation, an applicant’s responses were aided by an AI chatbot that misunderstood questions, providing answers that seemed disconnected or nonsensical. This miscommunication revealed the reliance on scripted responses and AI assistance, which, in a real-world scenario, could be a red flag.