We do not use artificial intelligence (AI) to make enforcement decisions at any stage of our process. Automated tools may be used for discovery and data collection, but all verification, assessment, submission, and reporting actions are carried out manually.
All enforcement decisions are based on verified evidence and human judgement. Each case is assessed individually to ensure that any action taken is accurate, proportionate, and defensible.
AI-driven systems are prone to inaccuracies and false positives. In a content protection context, this creates risk as DMCA notices are legal documents, and inaccurate submissions can expose both us and our clients to legal liability, as well as cause reputational damage.
Manual verification allows us to distinguish between genuine infringements and misleading or malicious sources, such as scam links that claim to host infringing content but instead attempt to extract personal or financial information.
All identification, verification, and submission work is carried out using OSINT methods and industry experience to ensure accurate, proportionate, and defensible enforcement. Each enforcement action is reviewed and submitted with clear accountability, ensuring that decisions can be justified if challenged by platforms, rights holders, or third parties.
We are experienced with AI tools, platforms, and AI-generated content ecosystems, and regularly carry out enforcement against infringements involving synthetic or AI-generated media. However, we do not rely on AI to make enforcement decisions.