This lawsuit was briefly mentioned in last week’s news of the week and has particular relevance to what we’re currently covering in class. Getty images is currently suing Stability AI, an AI based image generator, alleging they copied 12 million images belonging to Getty to train it’s AI model.
This case is particularly interesting to me as Getty is taking particular issue with the actual ingestion of copyright images used to train the AI model. To use an admittedly, potentially weak analogy, don’t we as students ‘ingest’ copyrighted materials everyday to produce our own work? Whether it be textbooks, lecture slides or CANs, we digest these copyrighted materials to produce our own notes; can it be said Stability’s AI is doing the same, albeit on a much grander scale?
Of course the question of how similar our own notes are to the said ‘ingested’ work arises, and I’m sure this will be an issue in the case in regards to similarity between stability’s AI generated images and the source material it’s derived from. But I question how the court will deal with the actual ‘ingestion’ itself and what the copyright implications of this might be.
Communications Law