I’m an Innocent Victim of AI-Based Algorithms
A personal case study of my Kafkaesque Instagram experience
Over the past three months, one of my two Instagram accounts — the main one where I post most of my art — has been the victim of the platform’s new AI-based tools designed to identify and root out bots to “protect the community.”
The first time I got ensnared was an attempted comment opening with “Srsly?” in response to what struck me as an unsupported statement touted in an ad. Twice over the next week, my comments were deleted. Then I was banned from commenting for a week, after being presented with a pop-up warning me, and providing a button to “Tell Us” if they made a mistake. The button is unclickable.
I did an online search to try to figure out what was going on, and read tips on how to not be seen as a bot. I did none of the things identified as possible triggers. I also discovered multiple discussion threads on Reddit of the same comment ban happening to other people.
I uninstalled and reinstalled Instagram on my phone. I made sure I had the latest version for my phone model. I switched from a business account to a personal one. I edited my profile to change the text, all in an attempt to signal that I was not a bot. Things were fine for a couple of weeks. I was able to post and to…
