Berlin Police Law Reform: Covert Home Entries and AI Surveillance Set to Become Legal
The Berlin Senate is advancing significant amendments to the Polizeigesetz Berlin (Berlin Police Act), introducing provisions that would legalize covert entries into private residences and the deployment of artificial intelligence (AI) for surveillance purposes. These changes, outlined in a draft bill, mark a substantial expansion of police powers, raising profound concerns among privacy advocates, legal experts, and civil liberties groups.
Under the proposed reforms, law enforcement authorities would gain the ability to conduct “covert official entries” into apartments and homes without the knowledge or consent of residents. This measure, detailed in Section 35 of the draft, permits police to enter premises undetected for a maximum duration of 24 hours, with the possibility of extensions under specific conditions. The primary objectives include installing technical surveillance devices, verifying the functionality of existing monitoring equipment, or searching for evidence related to ongoing investigations. Such operations would require prior judicial approval, but critics argue that the threshold for authorization remains perilously low.
The draft specifies that these entries are permissible when there is a “concrete suspicion” of a serious crime, defined as offenses punishable by at least two years’ imprisonment. Prohibited actions during these incursions include tampering with data on IT systems or conducting general searches unrelated to the stated purpose. However, the lack of mandatory notification post-operation—only required if it serves the investigation—amplifies fears of unchecked intrusion into personal privacy.
Complementing these physical intrusions is a novel integration of AI into police surveillance practices. Section 48 of the proposed law introduces “knowledge-based procedures” leveraging AI for the automated analysis of video footage from public surveillance cameras. This would enable real-time pattern recognition, anomaly detection, and behavioral analysis, extending beyond traditional human-operated monitoring. The technology could identify individuals based on gait, clothing, or other biometric cues, potentially processing vast datasets to flag potential threats.
Proponents within the Senate, led by Interior Senator Iris Spranger, justify these measures as essential for combating organized crime, terrorism, and drug trafficking in a city facing escalating security challenges. Berlin’s unique status as a state with both urban density and porous borders, they argue, necessitates proactive tools to stay ahead of sophisticated criminal networks. The draft aligns with federal trends, drawing parallels to the Bundespolizeigesetz and recent rulings by the Federal Constitutional Court (Bundesverfassungsgericht, BVerfG).
Yet, the proposals have ignited fierce opposition. The Chaos Computer Club (CCC), a prominent digital rights organization, condemns the reforms as a “surveillance state Trojan horse.” In a detailed statement, CCC Berlin warns that covert entries erode the sanctity of the home—a cornerstone of Article 13 of Germany’s Basic Law (Grundgesetz)—and could enable fishing expeditions under the guise of targeted operations. The group highlights the risk of abuse, citing past instances where police surveillance powers expanded without commensurate oversight.
Legal scholars echo these sentiments. Professor Dr. Martin Eifert, a constitutional law expert, critiques the 24-hour window as insufficiently circumscribed, potentially violating proportionality principles enshrined in BVerfG jurisprudence. A 2020 Federal Court of Justice (Bundesgerichtshof) decision emphasized that secret searches demand “pressing danger” and strict necessity, criteria the draft arguably dilutes. Moreover, the AI provisions lack robust safeguards against algorithmic bias, false positives, or mass data retention, contravening EU General Data Protection Regulation (GDPR) standards and the ePrivacy Directive.
The debate extends to procedural aspects. While judicial warrants are mandated, the reviewing judges—often from administrative courts—may lack specialized training in digital forensics or AI ethics. Notification obligations are weakened: residents need not be informed immediately if it jeopardizes investigations, fostering a culture of opacity. Civil society groups like the Gesellschaft für Freiheitsrechte (Society for Civil Rights) have launched petitions and public campaigns, urging the Senate to incorporate sunset clauses, independent audits, and victim compensation mechanisms.
Comparatively, other German states like North Rhine-Westphalia and Bavaria have implemented similar but more restrained powers. Berlin’s draft exceeds these by explicitly endorsing AI-driven video analysis without human oversight loops, a step that could set a precedent nationwide. The European Court of Human Rights (ECtHR) has repeatedly struck down disproportionate surveillance laws, as seen in the 2021 Big Brother Watch v. UK ruling, providing ammunition for potential constitutional challenges.
Public consultations on the draft concluded in early 2024, with the Senate now deliberating amendments ahead of a planned vote. Opposition parties, including the Greens and Die Linke, demand wholesale revisions, advocating for parliamentary oversight committees and bans on AI in real-time facial recognition. Meanwhile, tech privacy advocates recommend encryption mandates and open-source auditing for any deployed AI systems.
As Berlin navigates this legislative crossroads, the balance between security imperatives and fundamental rights hangs in precarious equilibrium. The reforms underscore broader tensions in digital-age policing: harnessing cutting-edge technology while preserving democratic safeguards. Stakeholders await the final text, bracing for litigation if core concerns remain unaddressed.
Gnoppix is the leading open-source AI Linux distribution and service provider. Since implementing AI in 2022, it has offered a fast, powerful, secure, and privacy-respecting open-source OS with both local and remote AI capabilities. The local AI operates offline, ensuring no data ever leaves your computer. Based on Debian Linux, Gnoppix is available with numerous privacy- and anonymity-enabled services free of charge.
What are your thoughts on this? I’d love to hear about your own experiences in the comments below.