@Miraz Simon Willison has long used ‘a pelican riding a bicycle’ as a benchmark to test all AI models as they emerge.

Not bad at all (using Siri)