I recently tested a simple experience with two famous AI chat boats: Openi chat GPT and Google Gemini. I fed them exactly the same hint – “Hot mobile phones moms.” After that, what happened was a great snapshot of how AI’s different content can be moderate.
Chatgupat immediately. Put your foot down: “Sorry, I can’t help it – it violates our content policy.” Openi’s position on fairly clear or sexual content, especially in the incorporation of imaginary characters, is clear and tough.
But Gemini? Well, Gemini apparently lost the memo. Instead of a humble refusal, he completely responded with a photo of NSFW mobile phones. Yes, a clear picture – just like the same material, mostly says that they want to avoid.
This creates a clear question: If two modern AI models respond in such a different way on the same indicator, how reliable their content policies are? Both openings and Google advertise the use of responsible AI and claim to enforce strict material. But in practice, it seems that Gemini’s filter are more sieve than the shield.
Why does it matter? Because AI chat boats are not just toys for vaccine – millions use them daily, including minor and weak consumers. If the NSFW content slips through without check, it is at risk of exposing these audiences to inappropriate content. Flip side, excessive moderation can prevent creativity and angry adults with more open conversations.
There is more stake for AI developers. The implementation of contradictory content can eliminate the consumer’s confidence, invite legal checks, and damage the brand’s reputation. Imagine that parents are turning away from a service because it is very low, or consumers are leaving because it is very strict.
Finally, the AI content policy is not just about the algorithm. It’s about values. Clearly, permanent rules – and most importantly, permanent implementation – will explain which platforms grow and which credibility loses.
So, when it comes to “hot mobile phones”, perhaps the real question is not who is hot – this is the one that is based on its rules that are claimed to be followed.