The providers of general-purpose AI models with systemic risks should also adequately identify, prevent and mitigate emerging systemic risks. Providers should keep up to date their understanding of the potential risks that may affect the general-purpose AI model they provide, and, as appropriate, periodically re-evaluate the risks, for example, after each substantive update of the general-purpose AI model. Providers of general-purpose AI models with systemic risk should also put in place appropriate mitigation measures without undue delay where necessary to address any new systemic risks that the provider becomes aware of. Providers should report incidents to the AI Office without undue delay after becoming aware thereof, as appropriate and without prejudice to any other reporting or engagement obligations under Union or national law, including information about new-found systemic risks and any corrective measures taken.