Series 13 of 16
Benchmarks and Evaluation
How progress is measured, and the gap to your experience.
Choose how to learn
Episode 1 · comingAI benchmarksWhat does 'state of the art' actually measure?
Episode 2 · comingBenchmark contaminationWhat if the test was in the training data?
Episode 3 · comingCapability versus reliabilityWhy does 'it can' not mean 'it will'?
Episode 4 · comingArtificial general intelligenceWhat would it mean, and who benefits from the claim?
Next seriesAI Infrastructure