OpenAI has put forth a compelling argument for the integration of AI in content moderation, highlighting its potential to enhance efficiency for social media platforms by expediting the handling of challenging tasks.
While generative AI has garnered substantial attention, major players like Microsoft and Google’s parent company Alphabet are yet to translate their significant investments into widespread industry impacts.
OpenAI, supported by Microsoft, revealed that its latest AI model, GPT-4, can remarkably shrink the content moderation process from months to just a few hours, ensuring more consistent content labeling.
Content moderation poses a strenuous challenge for social media giants such as Meta, the parent company of Facebook. These platforms rely on thousands of moderators worldwide to screen and prevent users from encountering harmful content like child abuse material and extreme violence images.
OpenAI emphasized, “The pace of content moderation inherently lags and can impose mental strain on human moderators.”
“With the introduction of this system, the arduous process of formulating and adapting content policies is condensed from months to mere hours.”
Furthermore, OpenAI’s CEO, Sam Altman, clarified that the startup does not train its AI models using data generated by users.
