Artists Revolt Against OpenAI Over Unpaid Contributions and Leak Sora's API Access
1min read Amid high expectations since its February unveiling, OpenAI's video-generating AI model, Sora, has yet to see an official release. On November 26, a group of artists involved in testing Sora reportedly leaked access credentials in protest against what they perceive as OpenAI's exploitative practices.
The API leak appeared on the AI model community platform Hugging Face, enabling users to generate videos from text descriptions for a brief period. This sudden exposure allowed the public to create high-resolution videos, quickly amassing 87 generated videos within just three hours before OpenAI revoked access to the early testing group.
OpenAI has not officially confirmed the breach, emphasizing that participation in their "research preview" was voluntary, with no obligation for feedback or tool usage. This incident has spotlighted transparency issues within the AI industry, prompting the artists to circulate an open letter for public endorsement.
These artists criticized OpenAI for leveraging their unpaid work for marketing gains without proper compensation. They likened OpenAI to "medieval feudal lords," exploiting artists' contributions while offering negligible rewards. Several artists expressed frustration over having to receive OpenAI's approval to share their creative works, feeling that the early access program prioritized public relations over artistic expression.
The reaction culminated in the release of an angry public letter comparing the company's tactics to a "whitewash," suggesting they were more about PR than supporting creativity. The artists urged their peers to adopt open-source video tools to evade what they see as a monopolistic grip on artistic innovation.
Following these events, OpenAI quickly cut off access to Sora, which remains in a research preview. They expressed gratitude for the artists' voluntary participation, promising continued support through grants, events, and other initiatives, underscoring a belief in AI as a formidable creative tool.
This episode illustrates ongoing concerns about transparency in AI development. The industry's practice of tightly controlling feedback from early testers often prevents independent scrutiny. OpenAI's situation highlights the delicate balance between innovation and ethical use, a topic increasingly scrutinized by AI safety advocates.