Evaluating ERM or GRC software? Get free Excel and Word RFP templates — 250+ requirements

Get the templates

Article 55(1)(a)

CurrentConditional timing

Evaluate and Adversarially Test GPAI Models With Systemic Risk

Applies to GPAI Provider; Systemic-Risk GPAI.

Actors: GPAI Provider
AI class: GPAI with Systemic Risk
Themes: Risk & Assurance

Tracker Networks Guidance

Perform model evaluation using state-of-the-art standardised protocols and tools, including conducting and documenting adversarial testing to identify and mitigate systemic risks.

Official text

Article 55(1)(a)Official source

(a) perform model evaluation in accordance with standardised protocols and tools reflecting the state of the art, including conducting and documenting adversarial testing of the model with a view to identifying and mitigating systemic risks;

Timing depends on the system

2 Aug 2025 — GPAI model placed on the market on or after 2025-08-02
2 Aug 2027 — GPAI model placed on the market before 2025-08-02

Suggested controls

GPAI Systemic Risk Evaluation and Adversarial Testing
Evaluate systemic-risk GPAI models using state-of-the-art protocols and adversarial testing.

Related risks

Unmanaged GPAI Systemic Risk
A systemic-risk GPAI model may create material Union-level risks that are not adequately evaluated, identified, monitored or mitigated.

PreviousArticle 54(3)-(5)NextArticle 55(1)(b)

Recognized by G2 as a Leader

Trusted by customers and rated highly across all categories

Article 55(1)(a): Evaluate and Adversarially Test GPAI Models With Systemic Risk | Tracker Networks