[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"framework:eu-ai-act\u002F14":3},{"item":4,"lessons":16,"controls":26,"related":40},{"family":5,"item":6,"code":7,"title":8,"citations":9,"publisher":10,"summary_md":11,"official_text":12,"official_attribution":12,"official_url":13,"link_label":14,"note":12,"updated_at":15},"eu-ai-act","14","Article 14","Human oversight",2,"European Union","**What it asks for.** High-risk AI systems must be designed so that natural persons can effectively oversee them during use, with the aim of preventing or minimising risks to health, safety and fundamental rights. Oversight measures are either built into the system by the provider or identified by the provider for the deployer to implement.\n\n**In practice.** The people assigned to oversight must be able to understand the system's capabilities and limitations, monitor its operation, remain aware of automation bias, interpret the output correctly, decide not to use it or override or reverse its output, and stop it safely, for example with a stop button. For remote biometric identification, the article generally requires that a result be separately verified by at least two competent people before action is taken. Evidence includes oversight procedures, override records, training for reviewers and interface designs that show confidence or explanation information.\n\n**Common gaps.** Oversight is often nominal: reviewers approve outputs in bulk without the time, information or authority to challenge them, and override decisions are not logged.",null,"https:\u002F\u002Feur-lex.europa.eu\u002Feli\u002Freg\u002F2024\u002F1689\u002Foj","Read the AI Act on EUR-Lex","2026-10-18T15:18:43.664+00:00",{"total":9,"list":17},[18,22],{"slug":19,"title":20,"published_at":21},"ai-model-autonomously-attempts-supply-chain-attacks-in-safety-testing","AI Model Autonomously Attempts Supply Chain Attacks in Safety Testing","2026-09-04T22:21:02.98+00:00",{"slug":23,"title":24,"published_at":25},"ai-models-struggle-with-complex-malware-analysis-human-oversight-remains-critical","AI Models Struggle With Complex Malware Analysis, Human Oversight Remains Critical","2026-07-23T14:20:52.894+00:00",[27,31,36],{"slug":28,"name":29,"problem":30,"stories":9},"ai-security","AI application security","Prompt injection, data leaks through AI tools and unapproved AI use get caught.",{"slug":32,"name":33,"problem":34,"stories":35},"software-supply-chain","Software supply chain security","Malicious and vulnerable open-source packages are caught before they ship.",1,{"slug":37,"name":38,"problem":39,"stories":35},"threat-intelligence","Threat intelligence","Attack campaigns reach your sector, suppliers or software before your team hears about them.",[41,46,51,55,59,63,67,72],{"family":42,"item":43,"code":44,"title":45,"together":9},"nist-ai-rmf","overview","AI RMF 1.0","AI Risk Management Framework (AI RMF 1.0)",{"family":47,"item":48,"code":49,"title":50,"together":35},"cis","16","Control 16","Application Software Security",{"family":47,"item":52,"code":53,"title":54,"together":35},"3","Control 3","Data Protection",{"family":47,"item":56,"code":57,"title":58,"together":35},"6","Control 6","Access Control Management",{"family":47,"item":60,"code":61,"title":62,"together":35},"8","Control 8","Audit Log Management",{"family":5,"item":64,"code":65,"title":66,"together":35},"9","Article 9","Risk management system",{"family":68,"item":69,"code":70,"title":71,"together":35},"iso","42001","ISO\u002FIEC 42001","ISO\u002FIEC 42001: Artificial intelligence management system",{"family":73,"item":74,"code":75,"title":76,"together":35},"mitre-atlas","aml-t0043","AML.T0043","Craft Adversarial Data"]