Instagram’s AI Content Label is stupid
Photo Courtesy of Airam Dato- Pexels
As the newly appointed PR Manager for The Lamron, I have the lovely opportunity to create content for this newspaper's Instagram page. This includes creating fun reels, article of the week recaps, and silly little posts I design through Canva.
I would consider myself a Canva expert. Give me four hours, a good playlist, and an iced chai and I will lock in to create the most aesthetic Instagram post this campus has ever seen. My last post was no exception. I spent three and a half hours working on a ‘Meet the E-Board’ post. I was so proud of it. I did a little searching on Pinterest for some template inspiration and a theme for the post. I landed on a junk journal theme, so I gathered everyone's information and created a lovely, well-put together, and informational post.
On Monday, Aug. 31, I uploaded the post to Instagram. I then went and shared it on my personal Instagram story, feeling excited for my friends and peers to check it out. That was until I refreshed my screen and saw that Instagram labeled the post as “AI Content.” At first, I was confused. I assumed it must have been some mistake. I refreshed the screen again, but sure enough it was still labeled as “AI Content.”
I was furious. I know what an AI flyer looks like when I see one. They all have the same font and an overused ugly style to them. I thought to myself, “why on earth would Instagram think my work was AI generated?” So, I clicked on the words “AI Content” to learn more. According to Instagram, the “signal typically identifies content that was entirely AI-generated instead of content that was only modified with AI.”
This led me to do some research; I found out that, according to Medium, “there’s a standard called C2PA that gets embedded when you use certain editing tools. The platforms [Instagram] read it and automatically apply the label.” Yes, Canva is an editing tool, but many of the AI featured tools are only accessible from the Canva Pro version. I do not have a subscription to Canva Pro. Now, the free version of Canva does have a select few AI tools for users, like the Magic Write feature, but I do not utilize those features as I find them entirely unhelpful.
My work and design for the Instagram post is not the same as some AI generated slop that was created in 10 seconds. I find it offensive and wrong that my post was labeled as “AI Content.” Additionally, I found this further angering as many flyers posted on the other campus organizations social media pages—that clearly use AI— were not flagged as “AI Content.” I do not understand why something so obviously AI generated is not being flagged, but a post that was created using an editing app that has AI features available is being flagged.
Do not get me wrong, I think it is incredibly important and necessary to have a warning on social media posts that contain AI, especially when it comes to art, news, and anything that could contain harmful content. These warnings would allow people to develop AI media literacy, something so many people lack. But, there needs to be a more efficient way to go about it. Maybe a good way to start could be having actual human beings help detect and report AI generated content, and not just another machine.