After a bit of a kerfuffle last week, with folks worried that Adobe's Terms of Service (ToS) might let them do just that, Adobe stepped up to set the record straight.

The ToS sent to Creative Cloud Suite users got everyone's knickers in a twist. It said Adobe "may access, view, or listen to your Content through both automated and manual methods -- using techniques such as machine learning to improve our Services and Software and the user experience."

Starting from 18 June 18, Adobe will introduce new ToS that clearly outline what they can and can't do with the content you create.

Adobe Digital media pooh-bah David Wadhwani said: "We have never used our customer's content to train generative AI, we have never claimed ownership of a customer's work, and we have never granted access to customer content beyond what's legally required."

Adobe's supreme product boss, Scott Belsky, admitted the wording of the ToS was a bit murky and stressed that there is not enough trust and transparency these days.

ToS language wasn't ever meant to let AI training happen on customer work, he said.

 "Looking back, we should've spruced up and spelled out the ToS a bit earlier," he said. We should've been more on the ball with narrowing the terms to fit what we actually do and explaining our legal must-dos."

Wadhwani is pleased with the process of moderating content for Adobe stock and Firefly training data, but he's the first to say it's not foolproof.

 He mentioned that Adobe can yank content that breaks the rules from Firefly's training data, and if you're not keen on automated systems that help spruce up their service, you can give it a miss.

Adobe's blog post says they get that trust isn't handed out like sweets; they've got to earn it. They're all ears for feedback on the new changes.