"those options range from "option 0", simply doing nothing and leaving UK copyright legislation in its currently uncertain state when it comes to the use of copyright materials to train AI models, through to options which would either require specific consent from rights holders in all cases ("option 1") or allow consent to be assumed by AI developers unless a rights holder objects, subject to developers being transparent about what materials have been used in training ("option 3")."
"The publication has revealed that over 11,500 responses were received (an usually high number) and that of the circa 10,000 responses submitted via the government's online "Citizen Space" portal, 88% of those supported option 1, with more of those respondents supporting option 0 of simply doing nothing (7%) than the government's preferred option 3 which was supported by a mere 3% of those respondents."
Government launched a February 2025 consultation seeking views on four options for how UK copyright should treat use of copyrighted materials to train AI models. The options ranged from option 0 (do nothing) through option 1 (require specific consent in all cases) to option 3 (assumed consent unless rights holders opt out, with developers transparent about training materials). The government indicated a preference for option 3 to balance rights holder and developer interests. Many high-profile rights holders campaigned against assumed-consent changes. A December 2025 progress update showed over 11,500 responses, with circa 10,000 via Citizen Space: 88% supported option 1, 7% option 0, and 3% option 3.
Read at Global IP & Technology Law Blog
Unable to calculate read time
Collection
[
|
...
]