ADDITIONAL AI TRANSPARENCY REQUIREMENTS FOR THE IMPLEMENTATION OF THE RELEVANT PROVISIONS OF THE ARTIFICIAL INTELLIGENCE ACT BY THE EU AS OF 2 AUGUST

On 31 July, the European Commission announced on 31 July that, as of 2 August, the Office of Artificial Intelligence of the European Commission would work with the competent authorities of member States to implement the relevant provisions of the Act. At the same time, the new artificial intelligence transparency requirements will also come into effect, requiring that some artificial intelligence systems clearly inform users that they are interacting with artificial intelligence and whether the content is generated or modified by artificial intelligence. According to the new rules, interactive artificial intelligence systems such as chat robots must clearly identify their artificial intelligence identity to the user, not to the real person. The contents of images, videos and audio, such as those generated or edited in depth by artificial intelligence, must be marked and those generated or modified by artificial intelligence must also be marked with machine-identifiable markings for identification and tracking. The European Commission indicated that these measures are aimed at reducing fraud and manipulation and at helping the public to make more informed judgements while providing enterprises with clearer compliance requirements and implementation pathways. On the same day, the European Commission also published a list of more than 180 institutions that signed the Code of Conduct on Transparency in Artificial Intelligence Generation. The guidelines are intended to facilitate the implementation of the relevant provisions of the Artificial Intelligence Act on transparency in the content of artificial intelligence generation。