Replication vs Failover Groups

Claro's US primary account experiences an unplanned outage at 02:00 UTC. The last scheduled refresh of the EU-West secondary account completed at 01:45 UTC. Claro's platform team promotes the EU-West secondary to primary using a failover operation. By 02:10 UTC the EU account is live and taking writes.

What is true about the state of data in the newly promoted EU account?