The developers of ChatGPT, OpenAI, claim that their engine can replace human content censors with a high degree of accuracy, more consistency, and without the emotional toll that comes with being forced to watch violent and abusive content for extended periods of time.
According to OpenAI, it has been utilizing the content moderation system it created, which is based on its most recent GPT-4 model, and has discovered that it is superior to a moderator with minimal training, though not as successful as the most experienced human moderators.
Even before generative AI, content filtering has been a major concern. The difficulties could be made worse by the new technology, which makes it increasingly simpler to provide false information and other undesirable content.