It doesn’t matter if the Netanyahu coffee video is real. We are screwed no matter what

Is Benjamin Netanyahu dead? According to this video posted on March 15 by the Israeli prime minister’s office, he’s alive and thriving. You may have seen it online, along with a rabid debate between the crowd who claims it is fake (it is not) and the people who say it is real (which is correct, as determined by fact checkers and independent intelligence analysts).

But we are not here to debate about what is true or not. What matters is the debate itself. It’s another point of proof in our new normal: Since AI can make up believable new realities, people now doubt reality itself, using that claim to support their beliefs and push their agendas.

The rumors of Netanyahu’s demise caught fire after the U.S. and Israel executed strikes on Iran on February 28. Following those attacks, the prime minister’s public appearances at military bases and targeted towns were heavily restricted, creating an information vacuum that Iranian state broadcasters eagerly filled with claims of his death. To squash the noise, his office dropped a casual clip on Telegram and X showing him grabbing a drink with an aide at The Sataf café on the outskirts of Jerusalem.

To mock the assassination conspiracists, as Reuters pointed out later, he leaned into a Hebrew linguistic pun where the slang for “dead” translates to being “crazy about” something. Referencing a past video in which bad compression made him look like he had six digits, he tells his aide: “I’m crazy about coffee. I’m crazy about my people. They are behaving phenomenally. Do you want to count my fingers? You can. Here and here. See?”

His bad puns and show-and-tell didn’t matter. Within minutes, the internet mobilized to declare the footage a forgery synthesized by artificial intelligence. Conspiracists pointed to the café’s cash register screen, falsely claiming it proved the footage was from 2024. 

“Got some serious questions about the validity of this blatantly obvious AI video…” one X user wrote.

“Close up blur shot which looks more AI than the last one. Even the clothes look sketchy,” said another. “Credit where due though, they gave some good prompt this time.”

Even Elon Musk’s Grok chatbot poured gasoline on the fire by confidently hallucinating and telling users the video was a “100% deepfake.”

A labyrinth of mirrors

But the Netanyahu footage is legitimate. Reuters confirmed that it was real by cross-referencing architectural details in the background with archival pictures of the establishment, as well as other photos and videos of the events. The verification team at Spanish national broadcaster RTVE ran the file itself through detection software like Google SynthID, Sightengine, and IVERES—all of which flagged the video as human-made.

Furthermore, slow-motion playback clearly shows the café’s register displaying the time as “14:59” on March 15, 2026. The coffee shop itself uploaded photos corroborating his presence, and Netanyahu followed up the next day with another video of him chatting with locals on the exact same patio. As open-source intelligence analyst Tal Hagin pointed out to RTVE: “This is not AI.” He says that “the edits are quite normal, especially in a video recorded for a prime minister’s social media. Almost all the errors can be explained by the cuts and angles.”

None of that evidence moves the needle anymore. This is like the case of the fake military influencer Jessica Foster, but reversed. With Foster, a million people willfully ignored obvious visual deformities because they desperately wanted her to exist. Here, people will instantly reject verified reality if it contradicts their ideological worldview. 

The perimeter of objective truth has been entirely vaporized by the computational power of silicon giants. We have essentially handed God-level reality-warping capabilities to any troll with a broadband connection and a few bucks to run the latest video-generation tools. Hagin summarized our current dystopia when those new models came out from China: “We are no longer at the stage where it’s six months away. We are already there: unable to identify what’s AI and what’s not.”

As we predicted three years ago, this “fun” technological parlor trick will become the ultimate political weapon and bullying wrench, and a huge legal loophole, allowing defense attorneys to casually dismiss damning security footage in actual crimes. On the flip side, it will be weaponized to digitally frame the innocent and the weak. 

I’m beginning to think that not even an aggressive educational campaign about generative AI will change the course of society. Silicon Valley and its Chinese brothers have gotten us into a labyrinth made of mirrors that show a mix of dreamed-up and real stuff. One that is packed with the Minotaurs of misinformation, injustice, and desperation. And the worst thing is that I don’t think there’s an exit. It feels like a new infinite plane of reality that leads nowhere but to oblivion—unless governments force the technology companies to put a stop to it. Fat chance that is going to happen.

View the full article