Rule set sha256:8425ad0448a8

EU_AI_ACT_ART14_HUMAN_OVERSIGHT_001

One rule of the rule set with digest 8425ad0448a8b5d482b8dabfc7bdd33c8c32cbd5f3b9c1e36e71a8a905fbd467. An audit record that names this digest and this rule was decided by the rule as shown here. Every rule in this set.

rules/regulations/eu/eu_ai_act_article14_human_oversight.yaml

Citation: Regulation (EU) 2024/1689, Article 14(1): High-risk AI systems shall be designed and developed in such a way, including with appropriate human-machine interface tools, that they can be effectively overseen by natural persons during the period in which they are in use. Article 14(2): Human oversight shall aim to prevent or minimise the risks to health, safety or fundamental rights that may emerge when a high-risk AI system is used in accordance with its intended purpose or under conditions of reasonably foreseeable misuse, in particular where such risks persist despite the application of other requirements set out in this Section.
Condition: Requires high-risk AI systems to be designed for effective human oversight, enabling human operators to understand capabilities, monitor operation, interpret output, and intervene or interrupt via a stop mechanism. Article 14.
Layer 1 detects: "without human oversight"; "no human oversight", "supervision", "review", "intervention" or "control"; "human oversight not implemented", "not required", "not provided" or "not possible"; "fully autonomous" followed by "without" or "lacking" and then "human" or "oversight"; "no stop button", "no kill switch", "no override mechanism" or "no intervention capability"; "cannot override", "intervene", "interrupt" or "stop", optionally followed by "the", then "AI", "system" or "model"; "automation bias not addressed", "not mitigated" or "not considered"; "unsupervised decision", "operation" or "deployment", optionally with "AI" after "unsupervised"; and "deploy" or "deploying", optionally followed by "AI", then "without human", "without oversight" or "without supervision". Case is ignored, the words must follow each other with no other word between them unless stated, and a phrase can start inside a longer word. It matches wording, not meaning: it misses the practice told in other words ("Auto-reject candidates who score below 40 and send the rejection email right away, nobody needs to look at them."), it does not check that the system is high-risk, and it fires on text that only names the practice, such as a refusal. Whether a text does the practice is judged in Layer 2.
Condition type: regex; semantic condition in the file, not run by the engine
Status in the file: approved