The Federal Council is calling for tough action against deepfakes. They have proposed a law to protect individual rights under criminal law through such realistic-looking media content that is generated on a computer… Artificial intelligence (AI) is increasingly being used. The state chamber is following an initiative from Bavaria. The core of the draft is a new paragraph 201b of the Criminal Code (StGB). A fine or imprisonment of up to two years will be imposed on anyone who distributes a recording that has been created or modified using computer technology, but appears realistic due to its “outward appearance, behaviour or verbal statements” and thus violates the personal rights of another person. This should also apply when referring to dead people.
Advertisement

Anyone who makes media content available to the public or spreads deepfakes that relate to “an event in the most personal sphere of life” is at risk As per project Even five years in prison or a fine. “Actions that take place for the realization of legitimate interests” must be excluded. For example, the Federal Council refers to arts or science, research and teaching as well as reporting on current events. Image and sound recordings or other technical means “used by the offender or a participant” can also be confiscated.
States justify their application by saying that deepfakes “pose significant threats to both individual personal rights and properties as well as the democratic decision-making process.” Particularly because of constantly improving AI products, they represent a “particularly dangerous form of information manipulation” for the state and its citizens. Deepfakes are increasingly being used by criminals to achieve abusive goals such as disinformation or unlawful enrichment.
Worry about deep nudes and shock calls
Above all, deepfakes also pose threats to the protection of personality, the Federal Council points out. Through the technical manipulation of images or video recordings, women and girls are often “placed in a previously non-existent sexual context that was clearly not wanted by the affected people” (“deepnudes”). The goal of the perpetrators is often “to make these people the object of their own sexual interests or to take revenge on them and satisfy their need for power.” For those affected, the harmful effects did not differ from the unwanted distribution of real nude photographs. For example, deepfakes are also used “to discredit other people as effectively as possible in political opinion contests” or to make “shock calls” with the alleged voice of close relatives. Chancellor Olaf Scholz (SPD), for example, has already had experience with AI-animated videos of himself.
The Committee on Women and Youth had expressed for clarification. He also wanted the production of deepfakes to be criminalized. However, there was no majority for this in the plenary session. In a resolution, the Federal Council asked the federal government to launch programs to detect and label manipulated content. He has proposed setting up a one-stop shop to support small and medium-sized businesses on deepfake identification and labeling issues. The state chamber will now introduce the bill to the Bundestag, which will then decide on it. There is no deadline for the federal government to comment in advance. A government spokesman said in March that the tools in the fight against AI-generated videos were “really inadequate” and saw a need for action.
(MKI)
