Dan Cohen (FCP)

A few weeks ago, someone pretending to be me reached out to one of Israel’s leading defense companies, claiming to be working on a sequel to my book “The Weapon Wizards”. The impersonator requested an interview with the company’s CEO, supposedly for the new book. Fortunately, the company’s spokesperson, whom I know personally, became suspicious. The phone number was unfamiliar, the email address was different, and something didn’t seem right. She reached out directly, and that’s how the impersonation was exposed. Then yesterday came another discovery. Since July, a YouTube channel called “Katz the Analyst” (good name, I have to admit) has been publishing videos featuring what appears to be me speaking from my home office. Except the videos are deepfakes, using real footage of me in my home office but replacing the voice and inserting entirely different content. The reach is negligible, but the messaging is intentional: Iran is portrayed as the smart, patient player in the Middle East; Israel is presented as a liability to the United States; Russia and China are depicted as the beneficiaries of the regional standoff. There are also conspiracy theories woven throughout. All of this has been reported to the relevant authorities in Israel, as well as to YouTube. But here is the bigger issue, and why this should concern everyone, especially with the Israeli elections just around the corner. We are already seeing AI-generated videos being used in election campaigns. Some are clearly labeled as AI-generated, while others are designed to deceive. The technology has reached a point where almost anyone can be made to say almost anything, with a level of realism that would have been unimaginable just a few years ago. Think about what this means in the final days before an election. A fabricated video of a candidate making an outrageous statement, a military official supposedly revealing classified information, or a journalist appearing to report a story that never happened could spread to hundreds of thousands of people before anyone has a chance to verify its authenticity. And even if the video is eventually exposed as fake, the damage might already have been done. There is also a second danger. As deepfakes become more common, genuine videos and authentic statements can increasingly be dismissed as fabrications. In other words, the technology doesn’t just make it easier to spread lies. It also makes it harder to establish the truth. I have no idea who is behind these impersonations or whether the incidents are connected. The limited reach of the YouTube channel suggests that, for now, the impact is minimal. But the implications are much bigger and should worry us all.