Whenever the automated individual decision-making under this Decision is used, the following safeguards and requirements shall apply:
the automated termination is based on predefined rules and objectively verifiable conditions;
the automated individual decision-making mechanism operates reliably, consistently and securely, minimises erroneous outcomes, and enables any systematic errors or disproportionate effects to be identified and addressed;
where technically feasible the automated termination is subject to a prior warning that allows the candidate to remedy the situation within a short period of time before the automated termination applies;
the basis and circumstances of the automated termination are recorded and reviewed;
the confidentiality, integrity and availability of relevant information and records of the automated individual decision-making is protected;
effective human intervention and review of the automated termination is provided;
the person carrying out the human intervention or review shall have the authority and competence to examine the relevant circumstances and to modify or reverse the automated decision;
the data subject shall have the right to express their point of view on the automated decision and to contest it.
The Office shall ensure that the technical and organisational measures used for automated decision-making comply with the safeguards laid down in paragraph 1.