If you haven’t caught the latest kerfuffle on the AI content collection front, let me give you the latest. And mark my words, this is only the beginning.
About a month ago Adobe changed their Terms of Service for their Creative Cloud products. The two areas of the new Terms that are drawing the most attention are sections ‘4.1 Content’ and ‘4.2 Licenses to Your Content’. Here are the offending sections
4.1 Content. “Content” means any text, information, communication, or material, such as audio files, video files, electronic documents, or images, that you upload, import into, embed for use by, or create using the Services and Software….
4.2 Licenses to Your Content. Solely for the purposes of operating or improving the Services and Software, you grant us a non-exclusive, worldwide, royalty free sublicensable license, to use, reproduce, publicly display, distribute, modify, create derivative works based on, publicly perform, and translate the Content….
This change has sent the Adobe user community into a firestorm causing Adobe to issue multiple modifications to their terms and make a floundering attempt to walk-back the changes and clarify that they never intended to steal your content. You only need to glance over their X feed to see that the damage control isn’t working.
Today we announced an update to our General Terms of Use for Creative Cloud and Document Cloud customers that makes the language easier to understand, including how we handle your content. Learn more on our website. https://t.co/7D0P8cltqx— Adobe (@Adobe) June 18, 2024
Their threads are now full of posts about Adobe replacement products and calls for the CEO to either make a public apology or to be fired immediately.
There are also threats of lawsuits which will allow litigants to gain access to internal emails and processes that could make the problems for Adobe grow exponentially. One June 17th the FTC announced it was taking action against Adobe and two of their senior execs for deceiving customers, early termination fees and inhibited cancellations. When it rains, it pours.
It’s not going to end here, in fact I think we’re just getting warmed up. Most of the large language solutions are trying to monetize content that has been ingested into their Large Language Models (LLM’s), the problem is that they don’t own that content, and I’m willing to bet that the majority of it, like this website, is copyrighted.
Microsoft charges $30 a month for Co-Pilot, it’s Gen AI solution that is now embedded in most of it’s products. However, one quick prompt of Co-Pilot reveals the flaw. In addition to the prompt response, Co-Pilot sites the source of the data it is returning and in almost every case, that data is protected by copyright. Microsoft has vowed to pay the legal costs of anyone who is sued for using content extracted from Co-Pilot, but who knows how long that promise will last.
The AI game is about to get a lot more interesting as we go through these legal machinations. I predict that content developers are going to find ways to prevent their data from being ingested into LLM’s, as well as adding explicit language into their Terms to prevent their content from being sold without compensation. This could be Napster all over again.