In a groundbreaking move that prioritizes creator rights and sets a new standard for the AI industry, Adobe has reportedly begun offering compensation to creators in exchange for licensing their videos to train its upcoming text-to-video AI tool. This revelation, brought to light by internal documents reviewed by Bloomberg, showcases Adobe’s commitment to developing advanced AI technologies while respecting the intellectual property of content creators.
According to the documents, Adobe is willing to pay creators $120 for videos that showcase people engaging in everyday activities, such as walking or expressing a range of emotions, including joy and anger. The company is also interested in acquiring anatomy shots of feet, hands, and eyes, as well as footage of individuals interacting with various objects like smartphones and fitness equipment. However, Adobe has made it clear that it will not accept copyrighted content, nudity, or any potentially offensive material.
The compensation offered by Adobe varies based on the usability of the video content, with rates ranging from $2.62 to $7.25 per minute of footage that will be utilized to train its AI model. This approach not only demonstrates Adobe’s respect for creators’ work but also ensures that the company’s training dataset will be diverse, representative, and ethically sourced.
Adobe’s decision to fairly compensate creators stands in stark contrast to the approach taken by OpenAI, the creator of the popular text-to-video AI tool, Sora. When Sora was first unveiled, it stunned the world with its ability to generate near-realistic video clips based on text prompts. However, when questioned about the training data used to develop Sora, OpenAI CTO Mira Murati remained evasive, raising concerns about potential copyright infringement and the use of videos without the consent of their creators.
While Murati eventually confirmed that OpenAI had used videos from Shutterstock to train Sora, she declined to comment on whether the company had also utilized clips from platforms like YouTube and Meta. This lack of transparency has led to speculation about the ethical implications of OpenAI’s training methods and has highlighted the need for greater accountability in the AI industry.
Adobe’s proactive approach to creator compensation not only sets a positive precedent for the development of AI technologies but also positions the company as a leader in responsible and ethical AI practices. By investing in the quality and integrity of its training data, Adobe is ensuring that its upcoming text-to-video AI tool will be built on a foundation of respect for creators and their intellectual property.
The details surrounding Adobe’s text-to-video AI tool remain largely under wraps, but the company has confirmed its development and plans to discuss it further later this year. With Adobe’s proven track record in developing cutting-edge genAI tools for image generation and editing, the addition of a text-to-video program to its portfolio seems like a natural and exciting progression.
As the competition to develop advanced AI tools intensifies, Adobe’s commitment to creator compensation and ethical AI practices sets a new benchmark for the industry. By taking the high road and respecting the work of content creators, Adobe is demonstrating that innovation and ethics can go hand in hand, paving the way for a future where AI technologies are developed responsibly and with the utmost regard for the rights of those who contribute to their creation.
