The business acknowledges that its previous policy was “too restricted,” in agreement with the Oversight Board.
“Made with AI” badges will soon be associated with a wider variety of films, music, and photos, according to Meta, which claims that its present method of classifying content that was generated by artificial intelligence is insufficiently specific. Beginning in the month of May, it will begin adding the label to media whenever it identifies AI image markers that are customary in the industry or whenever users agree that they are sharing content that was generated by software. Additionally, the label may be applied to items that are flagged by fact-checkers; nonetheless, it is highly likely that the firm will lower the ranking of content that has been discovered as being incorrect or manipulated.
Following a decision made by the Oversight Board over a video that had been deliberately manipulated to represent President Joe Biden touching his granddaughter in an inappropriate manner, the firm made the announcement of the measure. Due to the fact that the video did not breach the company’s policies surrounding modified media, the Oversight Board agreed with Meta’s decision to not remove it from Facebook. In spite of this, the board strongly recommended that Meta “reconsider this policy as soon as possible, given the number of elections that will take place in 2024.”
Meta states that it is in agreement with the board’s “recommendation that providing transparency and additional context is now the better way to address manipulated media and avoid the risk of unnecessarily restricting freedom of speech, so we’ll keep this content on our platforms so we can add labels and context.” In addition, the business has stated that beginning in July, it will no longer remove content only on the basis of violations of its policy regarding edited videos. “This timeline gives people time to understand the self-disclosure process before we stop removing the smaller subset of manipulated media,” wrote Monika Bickert, vice president of content standards at Meta, in a blog post to explain the timeline.
The label “Imagined with AI” had been applied by Meta to photorealistic photos that users had created with the use of the Meta AI tool. It is said by Meta that the revised policy goes beyond the suggestions made by the Oversight Board regarding labeling. “If we determine that digitally-created or altered images, video or audio create a particularly high risk of materially deceiving the public on a matter of importance, we may add a more prominent label so people have more information and context,” Bickert stated in his report.
Despite the fact that the firm is typically of the opinion that openness and allowing adequately labeled AI-generated photographs, images, and audio to remain on its platforms is the best way to move forward, it will nevertheless erase content that violates the guidelines. “We will remove content, regardless of whether it is created by AI or a person, if it violates our policies against voter interference, bullying and harassment, violence and incitement, or any other policy in our Community Standards,” Bickert explained to reporters.
Engadget received a statement from the Oversight Board expressing its satisfaction with the fact that Meta has taken its recommendations into consideration. Furthermore, it stated that it would conduct an evaluation of the company’s execution of these policies in a report on transparency at a later time.
“While it is always important to find ways to preserve freedom of expression while protecting against demonstrable offline harm, it is especially critical to do so in the context of such an important year for elections,” the board stated in its statement. As a result, we are thrilled to announce that Meta will start identifying a larger variety of video, audio, and image content as “Made with AI” when they find AI image markers or when people indicate that they have uploaded AI content. People will have access to more context and transparency for a wider variety of manipulated material, and posts that violate Meta’s rules in other ways will be removed as a result of this.