AI Generated Seinfeld-Like Sitcom Banned From Twitch For Transphobic Comedy

Twitch’s viral Seinfeld-like show generated by an AI has been banned. What happened all of a sudden?

AI-Seinfeld got banned on Twitch after transphobic jokes. | © Watch me Forever via Twitch

Using AI, "Nothing, Forever" can generate storylines featuring characters reminiscent of the hit 90s sitcom Seinfeld, as the main characters are parodies of their real-life counterparts.

The show was a hit on Twitch until ‘Larry’ – the show’s version of Jerry Seinfeld – has performed jokes that were deemed transphobic and homophobic, leading to a ban on the streaming platform.

AI-Generated Sitcom Banned On Twitch After Problematic Jokes

It all started with the segment that parodies the real Seinfeld’s opening, where Jerry does a stand-up routine in a club. Larry expressed himself in ways that were seen as transphobic and homophobic.

“I’m thinking about doing a bit about how being transgender is actually a mental illness. Or how all liberals are secretly gay and want to impose their will on everyone. Or something about how transgender people are ruining the fabric of society”

For Twitch, that was enough to cancel the show. The channel has been suspended for two weeks. The developers behind "Nothing, Forever" have commented on the incident, saying that this does not reflect their views and blaming issues with the AI.

On their Discord, they explained that changing the AI model seems to have had some negative side effects.

“We started having an outage using OpenAI’s GPT-3 Davinci model, which caused the show to exhibit errant behaviors (you may have seen empty rooms cycling through). OpenAI has a less sophisticated model, Curie, that was the predecessor to Davinci. When Davinci started failing, we switched over to Curie to try to keep the show running without any downtime. The switch to Curie was what resulted in the inappropriate text being generated”
Seinfeld 2
Larry is banned. | © Watch Me Forever via Twitch

The appearance of advanced AI on Twitch has prompted a lot of questions regarding the accountability of AI Twitch channels and how to handle the TOS surrounding that topic.

The creators of "Nothing, Forever" have apparently appealed the two-week Twitch ban, but we will have to wait and see, how the platform plans to manage AI streams. Will they be lenient, as it is a computer generated mistake? Or will they be strict, to set an example for future AI streamers to do better?