\n<\/p>\n
Meta has announced changes to its rules on AI-generated content and manipulated media following criticism from its Oversight Board. Starting next month, the company said, it will label a wider range of such content, including by applying a \u201cMade with AI\u201d badge to deepfakes. Additional contextual information may be shown when content has been manipulated in other ways that pose a high risk of deceiving the public on an important issue.<\/p>\n
The move could lead to the social networking giant labelling more pieces of content that have the potential to be misleading \u2014 important in a year of many elections taking place around the world. However, for deepfakes, Meta is only going to apply labels where the content in question has \u201cindustry standard AI image indicators,\u201d or where the uploader has disclosed it\u2019s AI-generated content.<\/p>\n
AI generated content that falls outside those bounds will, presumably, escape unlabelled.\u00a0<\/span><\/p>\n The policy change is also likely to lead to more AI-generated content and manipulated media remaining on Meta\u2019s platforms, since it\u2019s shifting to favor an approach focused on \u201cproviding transparency and additional context,\u201d as the \u201cbetter way to address this content\u201d (rather than removing manipulated media, given associated risks to free speech).<\/p>\n So, for AI-generated or otherwise manipulated media on Meta platforms like Facebook and Instagram, the playbook appears to be: more labels, fewer takedowns.<\/p>\n Meta said it will stop removing content solely on the basis of its current manipulated video policy in July, adding in a blog post published Friday that: \u201cThis timeline gives people time to understand the self-disclosure process before we stop removing the smaller subset of manipulated media.\u201d<\/p>\n The change of approach may be intended to respond to rising legal demands on Meta around content moderation and systemic risk, such as the European Union\u2019s Digital Services Act<\/a>. Since last August the EU law has applied a set of rules to its two main social networks that require Meta to walk a fine line between purging illegal content, mitigating systemic risks and protecting free speech. The bloc is also applying extra pressure on platforms ahead of elections<\/a> to the European Parliament this June, including urging tech giants to watermark deepfakes where technically feasible.<\/p>\n The upcoming US presidential election in November is also likely on Meta\u2019s mind.<\/p>\n Meta\u2019s advisory Board, which the tech giant funds but permits to run at arm\u2019s length, reviews a tiny percentage of its content moderation decisions but can also make policy recommendations. Meta is not bound to accept the Board\u2019s suggestions but in this instance it has agreed to amend its approach.<\/p>\n In a\u00a0blog post<\/a> published Friday, Monika Bickert, Meta\u2019s VP of content policy, the company said it\u2019s amending its policies on AI-generated content and manipulated media based on the Board\u2019s feedback. \u201cWe agree with the Oversight Board\u2019s argument that our existing approach is too narrow since it only covers videos that are created or altered by AI to make a person appear to say something they didn\u2019t say,\u201d she wrote.<\/p>\n Back in February<\/a>, the Oversight Board urged Meta to rethink its approach to AI-generated content after taking on the case of a doctored video of President Biden which had been edited to imply a sexual motive to a platonic kiss he gave his granddaughter. <\/span><\/p>\n While the <\/span>Board agreed with Meta\u2019s decision to leave the specific content up they attacked its policy on manipulated media as \u201cincoherent\u201d \u2014 pointing out, for example, that it only <\/span>applies to video created through AI, letting other fake content (such as more basically doctored video or audio) off the hook.\u00a0<\/span><\/p>\n Meta appears to have taken the critical feedback on board.<\/p>\n \u201cIn the last four years, and particularly in the last year, people have developed other kinds of realistic AI-generated content like audio and photos, and this technology is quickly evolving,\u201d Bickert wrote. \u201cAs the Board noted, it\u2019s equally important to address manipulation that shows a person doing something they didn\u2019t do.<\/p>\n \u201cThe Board also argued that we unnecessarily risk restricting freedom of expression when we remove manipulated media that does not otherwise violate our Community Standards. It recommended a \u2018less restrictive\u2019 approach to manipulated media like labels with context.\u201d<\/p>\n Earlier this year<\/a>, Meta announced it was working with others in the industry on developing common technical standards for identifying AI content<\/a>, including video and audio. It\u2019s leaning on that effort to expand labelling of synthetic media now.<\/p>\n \u201cOur \u2018Made with AI\u2019 labels on AI-generated video, audio and images will be based on our detection of industry-shared signals of AI images or people self-disclosing that they\u2019re uploading AI-generated content,\u201d said Bickert, noting the company already applies \u2018Imagined with AI\u2019 labels to photorealistic images created using its own Meta AI feature.<\/p>\n The expanded policy will cover \u201ca broader range of content in addition to the manipulated content that the Oversight Board recommended labeling\u201d, per Bickert.<\/p>\n \u201cIf 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,\u201d she wrote. \u201cThis overall approach gives people more information about the content so they can better assess it and so they will have context if they see the same content elsewhere.\u201d<\/p>\n Meta said it won\u2019t remove manipulated content \u2014 whether AI-based or otherwise doctored \u2014 unless<\/em> it violates other policies (such as voter interference, bullying and harassment, violence and incitement, or other Community Standards issues). Instead, as noted above, it may add \u201cinformational labels and context\u201d in certain scenarios of high public interest.<\/p>\nOversight Board criticism<\/h2>\n