“Placing on the market” is the point at which a model is first made available for use or distribution in the EU market. In practice, that can mean release through a product, API, license, or other channel that lets third parties access the model beyond internal development and testing.
This boundary matters because it helps तय? Avoid non-English. It matters because many EU AI Act duties attach only after the model is placed on the market. For cyber security teams, that line separates pre-release hardening from post-release compliance: threat modeling, red-teaming, access controls, and leakage testing happen before launch, while logging, incident triage, and reporting obligations can apply after launch. Attackers may target the release window to steal model weights, abuse exposed interfaces, or extract sensitive data before controls are fully in place. Defenders therefore track exactly when the model becomes externally available, since that moment can change both the security posture and the legal response.