OpenAI’s GPT Store Riddled With Spam And GPTs That Break Rules
The primary objective behind launching the GPT store was to provide a platform for developers to showcase their creations to a wider audience, while also enabling users to easily access a plethora of custom-made GPTs. However, the GPT store has encountered its fair share of hurdles, the media has reported.
According to a report by TechCrunch, OpenAI is currently grappling with issues such as spam and the presence of rule-breaking GPTs within the platform, posing challenges to its operation and management.
The TechCrunch report added that their investigation discovered that the GPT Store, designated as OpenAI’s official platform for GPTs, is inundated with peculiar and potentially copyright-infringing GPTs, suggesting a lax approach to moderation efforts by OpenAI. A quick search reveals GPTs claiming to produce artwork resembling Disney and Marvel properties, yet primarily function as conduits to third-party paid services.
Additionally, these GPTs boast about their ability to circumvent AI content detection tools like Turnitin and Copyleaks.
In order to list GPTs on the GPT Store, developers are required to verify their user profiles and submit their GPTs to OpenAI’s review system, which utilises a combination of human and automated review processes. Here is a statement from a spokesperson regarding the process, as quoted by the report:
“We use a combination of automated systems, human review and user reports to find and assess GPTs that potentially violate our policies. Violations can lead to actions against the content or your account, such as warnings, sharing restrictions or ineligibility for inclusion in GPT Store or monetisation.”
OpenAI’s Usage Policies explicitly forbid the utilisation of third-party content without the requisite permissions, thus prohibiting any instances of copyright infringement. Although OpenAI may not bear legal responsibility for these violations under the Digital Millennium Copyright Act, it is obligated to promptly remove any infringing content upon receiving a request to do so.
