I Couldn’t Believe My Eyes

-first published in Substack blog “The “Last Editor” on September 11, 2025

A brutal political assassination tells much about how far the country has descended into violence, but also spotlights our crisis of video believability.

I had just sat down to lunch yesterday when I got a text from my elder daughter with two words— “Holy Shit!” and a Reddit link. I followed the link, expecting to see an idiot driver pulling some dumb move on the road or other such video she often sends me. Instead, Reddit served up a video, recorded from a long distance away, of what appeared to be conservative activist Charlie Kirk at an outdoor rally. Suddenly, Kirk jerks back and the crowd begins to scramble. The video—now removed from Reddit—is hard to see, so just as I am about to search for the story, my daughter sends me another link, this time on Twitter/X. Recorded from much closer with sharp, clear video, I watch as what appears to be Kirk takes a shot to the neck and slumps over, blood spewing from the wound.

My first thought—that’s fake.

Many people are writing today about the decline in our ability to disagree peacefully about matters of politics and policy.

I don’t disagree with that sentiment. Political violence is abhorrent and has reached a level that is far beyond what we should accept. I certainly share those concerns, but my surprising reaction to a video that very much turned out not to be fake exposes another crisis that is just as threatening to our democracy. It has become far too difficult to tell real video from AI or other fakes, leaving journalism with the responsibility to try to figure out how to deliver only accurate videos and images for its audience.

This is, of course, just one of the many problems journalism now faces that are a threat to its very existence. It’s not lost on me that, 24 years ago today, as one of the biggest stories this country has ever seen unfolded, people turned to trusted journalism sources to see what was happening. The internet existed in 2001, but its strength in delivering news was in text and photos, not videos. So we turned to our televisions to see the video of the planes crashing into the towers, of the towers burning and eventually crashing to the ground. Those were incredible scenes of violence, but it was bloodless violence and we all aired it over and over again. AI and deepfakes didn’t exist then, so most of us took it for granted the videos we were watching were real. Interestingly, the nascent internet did help in the formation of the 9/11 truther movement, letting conspiracy theorists find each other and build their stories about what “actually happened” that day. Still, some minor conspiracy theories claimed the videos of the planes were faked to cover missile attacks or other causes for the collapses, perhaps an early warning for what we’re really facing now.

Journalism—real journalism—did ultimately help me decide the Kirk video was not fake.

As soon as I saw the close-up video, I went to the New York Times to see if it was covering the story. It was, with enough details at the point to say a shooting had happened. The Times was not showing either of the videos I had seen on social media, but I attributed that to the graphic nature of the footage more than to the videos being fake. Even as I write nearly 24 hours later, I’m not finding any reputable news outlet showing the graphic video. It is available, of course, in countless places on the internet, just not in established journalism outlets.

Some analyses I’ve already seen point to the lack of availability of that video on traditional news platforms as a problem for journalism, that the world has sidestepped us as gatekeepers. That has certainly become true in the internet age, so no startling revelations there. But it may also be a reason to keep journalism around. If we can find a method by which we confirm the veracity of videos and photos and can figure out a way to show that method to the public as something it can trust, then we could carve out a niche that would make us more valuable to some.

I’m not kidding myself to think there is some magic pill that will convince everyone what we do is truthful and worthwhile.

Many have gone too far down the path believing in “fake news” to ever find their way back. But for those reasoning news consumers, we can stake out our space. Perhaps it’s a “seal” of authenticity we develop jointly across media. I could see an organization like the National Press Photographers Association (NPPA) taking the lead on this, with the financial support of traditional media outlets. NPPA would hire a team of computer and visual experts and invest in the tools—again with our financial help—to quickly analyze images and videos for the telltale signs of AI or other manipulation.

Those that pass the test get the seal and go out to reputable news outlets supporting the process. That would allow the public to trust those images. It is not unlike what Snopes.com does now investigating stories, rumors, images and more that appear online. This proposed seal of approval harkens back to the old Underwriters Laboratories (now UL Solutions) seal that appear on electrical appliances, allowing consumers to know the device is tested and declares safe to use in the home. That seal was, and still is, wildly successful and easily recognized as to what it means. Ours could be the same. Sure, some people will never believe we are telling the truth. But let’s focus on those with whom we can still reason.

There’s no easy solution to the problem of fake videos and images and how much they undermine the confidence of the audience in the news we deliver. There clearly will always be reasons for someone to fake what we see online—political reasons, economic reasons, even entertainment reasons. But we can’t just shrug our shoulders and say that’s the way it is. A defeatist position like that would do more damage to the country than any fake video ever could.