Chris Fiumefreddo

FAQ for The Great Filter

Thank you for taking the time to look at The Great Filter - I hope it helps to reduce the anxiety inherent in trying to scroll through a social media feed these days.

I wanted to post an FAQ about the state of the tool right now:

Does it work on mobile? Not yet. The extension requires a desktop browser today. I have a Safari extension built and ready to test, but I don't have the resources to get it completely up and running. I really want to get this into the hands of mobile users, but I want to make sure it's the right approach and works just as well as the Chrome version. I hope to have it ready within a few months.

What about Facebook / Instagram? Testing it right now and hoping to have it ready for release soon. Facebook's DOM is more complex to work with reliably. I want the existing platforms to be rock solid before expanding.

How does the AI decide what to filter? We're using Claude (Anthropic's model) to evaluate each post against your chosen mode. The prompt is tuned to the harm patterns that are most common on each platform — outrage farming, financial manipulation, coordinated harassment. You can see the reason for every filtered post.

Can I see the reasoning? Yup, the only thing this tool is hiding is harmful content. Every placeholder shows why that specific post was filtered. If you really want to see it, you can just click on the View Anyway button. There's also your personal Intelligence Report showing category trends over time, so you can see what your corner of the internet has been serving you. That all resides on the user's end, I'm not even keeping broad category statistics. Personally, I've seen my report shift from my initial problems (grief farming, financial scams) to a more even distribution that makes me feel like I'm finally in control of what I get to actually engage with.

Will it work on this other social media site? If you have a social site that you want to see less mentally harmful content on, I would be happy to see if I can make the filter work on. Please feel free to submit a feedback comment through the website.

6 views

Add a comment

Replies

Best
Chris Fiumefreddo

Update: version 0.8 is now released, which will allow people to test it out without signing up for a free account.

For the planned 1.0 release, I’m hoping to have figured out Facebook and Threads filtering. Additionally, I’m working on a toggle switch that will allow users to just filter out AI generated content for those that don’t care about harmful posts but are tired of all of the obviously AI content in their feeds.

If you try it out, please let me know what you think.