Publisher policies on AI use: is it time for change?

What do you think – are current publisher policies on AI use robust enough to ensure research integrity?