Where do they get the video training data?
> Datasets. We construct a diverse and high-quality collection of video datasets to train STARFlow-V. Specifically, we leverage the high-quality subset of Panda (Chen et al., 2024b) mixed with an in-house stock video dataset, with a total number of 70M text-video pairs.
Wonder if "iCloud backups" would be counted as "stock video" there? ;)
They shared audio Siri recordings with contractors in 2019. It became opt-in only after backlash, similar to other privacy controversies.
This shows that they clearly prioritize not being sued or caught, which is slightly different from prioritizing user choices.