Twitter unveils design rules for deepfakes policy and asks for user feedback

Twitter has presented the first few draft rules that should shape the policy surrounding the action against deepfakes. In addition, a questionnaire has been put online so that users can influence how Twitter should deal with deepfake material.

Del Harvey, a Twitter vice president of trust and safety, describes deepfakes as “synthetic and manipulated media” and presents the option to act against it by posting a note next to tweets sharing such material. She also said that Twitter could choose to issue a warning if people like or share Twitter posts containing deepfakes. Another option is to add a link, where users can read more about why various sources believe that specific cases are ‘synthetic and manipulated media’.

Twitter only wants to act when it comes to this kind of content that deliberately misleads or confuses people. If a message containing deepfake material is misleading and could endanger someone’s personal safety or otherwise cause serious harm, Twitter suggests removing the message in its entirety.

These are design rules for now. Twitter already announced last month that it will ask for feedback from users in order to shape the policy. This feedback can be given in the form of a short questionnaire that has now been published. In it, for example, Twitter asks whether people prefer to remove or maintain deceptive messages with deepfake material, or whether they prefer a middle ground in the form of a warning label. This questionnaire will be online until 27 November.

Twitter won’t say when it will finalize its policy, but in any case, the company will make a new announcement 30 days before the policy goes into effect. Incidentally, a form is also available for people interested in helping Twitter develop solutions to detect synthetic and manipulated media.