Announcing: Riddance

A media literacy publication for the AI age

How we got here

AI-generated media moves too quickly for everyone. The social media companies don’t know what to do. No one has reliable detection solutions. Laws haven’t kept up. It feels like those responsible for AI media’s advancements are looking for others to deal with the damage left in the wake.

I started the ShowtoolsAI social channels in June of 2025 to talk about AI video. My objective was unclear at the time, but I believed AI-generated media could be a “tool” for “shows” in the best-case scenario, so long as it was used ethically. Funny enough, I registered the showtools.ai domain one day before Google’s Veo 3 came out. Oh, to be young and naïve…

I’ve been talking about how to spot AI videos since then, using the skills I developed as a live-streaming technical producer, camera operator, and sound engineer to discern real from AI. After building encoder tests and demonstrating camera shading methods, then producing hundreds of multi-cam live streams, I had (unfortunately) learned everything that could possibly go wrong with real videos. AI video looked like completely different — it might as well have been a video game.

As the gap between real and AI videos (and photos, music, text…) narrowed, our methods had to adapt.

Building a team and method

Mason Broxham texted me on July 3rd 2025, the day he left Microsoft. A friend since college, I knew him as an academic who studied linguistics and philosophy, who then transitioned into tech as a threat analyst specializing in AI detection. The need for AI literacy and detection was shifting. Once just an enterprise and cybersecurity concern, it was now essential for regular people to understand what was happening and what to do about it.

After a couple months of discussion, we started building. Mason got to work on an epistemic framework for tracking AI on social media, which turned into a database that stored AI videos and software to analyze and track them. Our goal was to capture the “fingerprints” of real and automated accounts, which turned into a full research methodology. Behind the scenes of my short video production schedule, we were building a system for long-term detection and tracking. We still are; the tools aren’t publicly available yet.

So, creating a place where we could share research, investigations, and educational material was the logical next step.

Why we’re launching Riddance

It’s weird to go from video to text while most journalists are doing the opposite, especially when your subject matter is other videos. But there are a ton of reasons to focus on writing:

• The AI world moves really fast. Short-form videos were my answer for a long time, and they will still be an essential part of our reporting. But when you need to get important information out quickly, writing is much faster.

• The short-form video format is incredible for discovery, but it's really bad for nuanced discussion about media literacy.

• Long-form videos have an expectation of being durable or “evergreen”. My “How to spot AI videos in 2025” is already outdated 6 months later and still finds new audiences. Leaving it up for much longer will start actively doing harm (it may be down when you see this).

• The video recommendation algorithms are a huge part of the AI problem and might not be long for an AI-infested world. Who wants to see new things if they’re probably fake? That’s a hot take for another time.

• Video platforms probably aren’t going to like what we have to say about them, nor will our reporting always be appropriate for video.

But most importantly, writing and reading are awesome! I get messages all the time from teachers who want written material, or people who wish their parents could read our media literacy tips rather than watching them in quick, short-form videos. And when you’re conscious about the problems with video feeds, it feels good to just write and not contribute to that noise.

Our goal is to grow a full team so we can do more investigations and educational work. There’s no shortage of topics to cover, and a ton of demand.

Why you should subscribe

We will provide:

• News about advancements in AI media

• Analysis and investigations on how people are using AI media

• Toolkits for media literacy tailored for the AI age

That’s the baseline. We’ll make as many posts freely available as possible an appreciate your subscription.

If you can afford to support us, your paid subscription gets:

• Occasional paid-only posts

• Access to the chat and community

• Full access to the archive

We’ll keep posting videos. Maybe we’ll experiment with some live streaming or podcasting. We’ll keep evolving and will appreciate your company along the way.