It also can't be used as a gatekeeper in that sense precisely because of its inherent fallibility. It will definitely miss some gen-AI writing and return false positives on real human writing. It's only supposed to be "much better than nothing" for people who reject the inhuman homogeneity of AI-writing and want to see less of it on their screen.

I'm not sure I follow your trickle-down argument, but in any case, the system is dynamic. It checks daily for new models and human-authored articles on their specific ticks, and cross-checks a corpus of known human-authored works. It will track both AI and human writing over time and adjust accordingly.

To the second point, that's the idea behind the adjustable threshold. It is indeed hard to distinguish; the extension will miss some gen-AI writing and return false-positives on some human writing. The user gets to decide which side to err toward.

It will never be infallible because the writing won't have C2PA metadata attached to it the way media does. It's just a tool that is a whole lot better than nothing for people who don't want to waste their time reading something written by a robot, or hate the inhuman homogeneity of "AI style." Don't you ever find yourself thinking "nobody talks like this?"

If you need people to take you seriously and there is a chance that they will judge you for your gen-AI content because they know it's not your own work, would you still post it?

That scoring methodology is the only part of the extension that's not open source; it's hidden behind a Cloudflare worker to prevent reverse engineering.

AI-speak won't become average-speak if I have anything to do with it. I want people to continue using their own brains to construct sentences rather than farming them out to machines.

For writing, it checks known vocab and punctuation tells, as well as subtler tells related to cadence, and assigns it a score subject to an adjustable threshold. If the text fails, users have the option to flag offending text, hide it, or block the page entirely (with the option to see anyway).

For media, it's admittedly fairly weak, as it relies on C2PA metadata which is stripped from all of the social media sites where it would be most helpful. (Anyone else have chronically online boomer parents continually gobbling up slop like it's real information?)

I have a D-slop+ version in the works that should be able to handle the media itself, but it's going to have to make API calls to have real teeth, which means I can't offer it for free. If this extension validates the concept, I'm happy to build it for y'all.

Yes, I vibe-coded it, but an ancillary bonus to the project accrued when it inspired me to cook dinner listening to Metallica's "Fight Fire with Fire," which in turn brought my 5 y/o running into the kitchen with every musical instrument in the house for an impromptu karaoke speed metal session.

It's MIT license open-source, full brief at https://github.com/jared-the-automator/d-slop; This forum is full of people smarter than me, so I'm open to suggestions.

Comments URL: https://news.ycombinator.com/item?id=48174775

Points: 10

# Comments: 11