> trounce human thumbnails (often nearly doubling the click-through rate according to A/B testing) That stat sounds more than dubious. Do you have a reference

> trounce human thumbnails (often nearly doubling the click-through rate according to A/B testing) That stat sounds more than dubious. Do you have a reference for it that isn't published by someone with a vested interest in people using AI?