Germany - Protection Against Deepfakes (222/24)

Overview

The Federal Council draft (BR-Drucksache 222/24) proposes a new §201b StGB titled "Violation of personality rights by digital falsification" to criminalise the misuse of digitally manufactured or altered media that give the false appearance of authentic images, audio or video of a person. The initiative — launched by the Free State of Bavaria and decided by the Bundesrat on 5 July 2024 — was transmitted to the Bundestag as BT-Drucksache 20/12605. The draft frames Deepfakes as a growing threat to private life, bodily and reputational integrity, property and democratic discourse and therefore seeks to create a tailored criminal-law response alongside civil-law remedies and platform-level responsibilities that exist under separate instruments. For the official submitted draft, see the Bundestag submission (BT-Drucksache 20/12605) and the Bundesrat decision record. BT-Drucksache 20/12605 (PDF). ([bundestag.de](https://www.bundestag.de/presse/hib/kurzmeldungen-1017140?utm_source=openai))

Definitions

The draft distinguishes "media content" as computer-generated or computer-altered visual or audio recordings that (1) imitate or recreate a person’s external appearance, behaviour or spoken utterances and (2) give the impression of a genuine, real-world recording. The statutory language targets those manipulations that are likely to mislead observers into believing the content documents authentic events or expressions of the depicted person. The draft further distinguishes between ordinary infringements and aggravated cases where the content is disseminated publicly or concerns the most intimate sphere of life (sexual acts etc.). The proposed text also covers cases concerning deceased persons if their personality rights are gravely affected by the falsification.

Governance and Institutional Framework

The principal institutional actors for the draft's pathway are the Bundesrat (initiator of BR-Drs. 222/24), the Bundestag (recipient; legislative decision-making body), and the Federal Ministry of Justice (Bundesministerium der Justiz) which prepares government positions and legislative processing. Law enforcement and prosecution authorities (Länder-level public prosecutors and police) would be responsible for investigating offences under a new §201b, and courts would adjudicate guilt and penalties. Separate administrative and civil-law avenues — including platform notice-and-takedown procedures under the Digital Services Act (DSA) implementation and civil personality-rights litigation — would operate in parallel, requiring coordination across ministries, prosecutors' offices and digital platform regulators. For the Bundesrat decision and subsequent government handling, see the official cabinet reference and Bundestag documents. Bundeskabinett: reference to BR-Drs. 222/24. ([bundesregierung.de](https://www.bundesregierung.de/breg-de/aktuelles/bundeskabinett-ergebnisse-2303574?utm_source=openai))

Key Focus Areas

The draft's core policy objectives are (a) to provide direct criminal-law protection for personality rights specifically tailored to harms from Deepfakes, (b) to deter malicious manipulation and distribution of hyper-realistic falsified media, (c) to create graduated sanctions reflecting seriousness (private circulation vs public dissemination and sexual/intimate content), and (d) to preserve space for legitimate uses by excluding socially legitimate purposes (art, science, research, reporting). The proposal responds to perceived gaps in existing law: for example, §201a StGB protects the highest private sphere but presupposes an underlying "image capture" of a real event, and traditional offences (defamation, insult, data-protection breaches) do not neatly capture the informational deception and contextual demeaning that Deepfakes produce. Critics emphasise risks of overbreadth, evidentiary difficulties (proving falsification and intent), freedom-of-expression trade-offs and overlaps with civil law and EU regulatory instruments. Several professional bodies including the Federal Bar Association have commented on drafting precision and enforcement concerns. Bavarian Justice Ministry press release. ([justiz.bayern.de](https://www.justiz.bayern.de/presse-und-medien/pressemitteilungen/archiv/2024/57.php?utm_source=openai))

Implementation Framework

If adopted, the new offence would require practical implementation steps across prosecutorial units (training to detect AI-manipulated media), digital forensics capacity building (expertise and tools to authenticate media), procedural adjustments (search, seizure and preservation of volatile AI artifacts), and inter-agency cooperation with platform regulators to ensure illegal material is removed swiftly from hosting services while preserving evidence for criminal prosecution. The draft itself provides for criminal sanctions and possible confiscation/seizure measures in serious cases; courts and prosecutors would need national guidelines and operational standards to apply the new provision consistently. Close coordination with EU-level actors would also be necessary for cross-border content and hosting, and for evidence-sharing. BT-Drucksache 20/12605 (PDF).

Monitoring and Evaluation

Effective monitoring would require metrics for enforcement (number of investigations, charges, convictions), removal efficacy on platforms, recidivism and measured impacts on victims (removal time, harm mitigation). Independent review mechanisms and periodic reporting (annual or biennial) by the Ministry of Justice and cooperating agencies would be recommended to evaluate whether the statutory design balances deterrence with fundamental freedoms and whether unintended chilling effects on legitimate expression occur. The Bundestag’s legislative committees and parliamentary oversight functions would play roles in evaluating outcomes and recommending amendments. ([bundestag.de](https://www.bundestag.de/presse/hib/kurzmeldungen-1017140?utm_source=openai))

Penalties, Liability, and Appeals

The draft sets out two principal penalty bands: (1) general cases with imprisonment up to two years or fines; (2) aggravated cases where the content is made public or concerns the most intimate sphere, carrying imprisonment up to five years or fines. The text contemplates confiscation or seizure of storage media used in the offence in serious situations. Criminal liability is directed at the person who makes the falsified media accessible to a third person; the draft's scope relative to intermediaries and platforms is less prescriptive and would depend on prosecutorial/practical approaches. Defendants would retain full procedural rights including appeal; courts must balance Article 5 (freedom of expression) and Article 1/2 (human dignity and personality rights) of the German Basic Law in contested cases. Commentary notes constitutional scrutiny risks if wording is overly vague.

Relationship to Other Instruments

The proposed §201b is designed to supplement — not wholly replace — existing civil and administrative avenues (injunctions, damages, platform regulatory duties). It intersects with data-protection law (DSGVO/BDSG) when Deepfakes use personal data, intellectual-property/portrait rights (KunstUrhG / KUG) and other criminal offences (defamation §187 StGB, insult §185 StGB, fraud §263 StGB where deception causes financial loss). EU-level instruments also interact: transparency obligations and other AI-specific provisions in the EU AI framework and the Digital Services Act affect platform responses to flagged content. Legislative drafters must ensure coherence with these instruments to avoid duplication or contradiction. ([bpb.de](https://www.bpb.de/lernen/bewegtbild-und-politische-bildung/556843/strafrecht-und-regulierung-von-deepfake-pornografie/?utm_source=openai))

International Alignment

The Bundesrat draft recognises that Deepfake regulation is an international challenge involving cross-border hosting, distributed authorship and differing criminal standards. The draft’s objectives align with EU-level policy trends seeking to address online sexual violence, disinformation and AI-specific risks; however, approaches vary between jurisdictions (some criminalise only sexual Deepfakes or apply platform liability on different grounds). Effective enforcement will depend on mutual legal assistance, cross-border content takedown cooperation, and alignment with EU acts (AI Act, DSA) to ensure platforms and service providers have clear obligations and mechanisms for evidence preservation and removal. International human-rights law and free-expression norms must be considered in any harmonisation effort. ([bpb.de](https://www.bpb.de/lernen/bewegtbild-und-politische-bildung/556843/strafrecht-und-regulierung-von-deepfake-pornografie/?utm_source=openai))

Implementation Timeline

Sources and References