AnalysisAI ModelsAugust 19, 2026

MIT study finds AI-generated images often can't be traced to training data

Read original source →news.mit.edu

MIT CSAIL researchers identify "attribution decay": the more data an image generator trains on, the less any single training image — or all images by one artist — affects outputs. Lead author Zheng Dai argues if deleting data doesn't change the output, it can't be attributed. David Gifford calls it the first method proving deleted inputs have zero influence.

People · Zheng Dai, David Gifford

1 source

More stories today

Open the live feed