Adobe has not done its reputation any favors these last few weeks and the latest story to break the peace is Adobe’s terms of service changes. We’ve said it before and we’ll likely have to reiterate this point many times over for a long time to come, but AI is everywhere, and how these models are to be trained to do their jobs is a moral gray area that is regularly debated. While tensions around AI are already high, Adobe announced a terms of service change in which the company “clarified that we may access your content through both automated and manual methods.” 

The TOS section that was updated stated that “techniques such as machine learning” could be used to analyze user data in order to improve their services. The terms did not specifically mention anything about an “AI” policy change, but users who saw Adobe’s user agreement change were immediately alerted to the possibility that all of their work on the platform could be used to train the company’s proprietary AI, Firefly. 

Adobe terms user backlash

Image: Adobe

Adobe Terms of Service Change Sparks Outrage Among Users

The Adobe TOS controversy has more to do with miscommunication than the actual use of any user data for AI training. Adobe’s terms generated user backlash as many felt the company was letting them know that their personal data was freely available for the company to access and do what they needed with it. This was not the case. The company’s policy change to access user content was not a new addition to their terms and similar statements to view and take action against inappropriate content have been a part of the policy for a long time. 

Adobe Addresses TOS Controversy To Clarify the Source of AI Training Material

In a blog post, the company expressed its stance on AI and clarified that it had no plans to use user data to train generative AI. “We’ve never trained generative AI on customer content, taken ownership of a customer’s work, or allowed access to customer content beyond legal requirements. Nor were we considering any of those practices as part of the recent Terms of Use update. That said, we agree that evolving our Terms of Use to reflect our commitments to our community is the right thing to do.” From the statement, it’s quite evident that the Adobe terms that led to user backlash resulted from a lack of clarity on their AI policy, intensified by their statement on accessing user data. 

Addressing the Adobe TOS controversy, the company reiterated the details of how their AI is trained, ensuring they made a clear statement on the datasets that were used, “Adobe Firefly is only trained on a dataset of licensed content with permission, such as Adobe Stock, and public domain content where copyright has expired.” This doesn’t explain why users have been able to use artists’ names as search terms to generate commercially available stock AI images that mimic their art style to a degree at least. 

More recently, photographer Ansel Adams’ estate called out the company’s use of “Ansel Adams-style,” stock images. If the AI isn’t trained on their art, then how does it know to create images that take references from their work? Many blame crowdsourced content for the error but Adobe claims to have a system in place to prevent this from happening.

 

Post by @anseladams
View on Threads

 

This point has not received as much attention as it should have, but as it stands, the company’s policy explicitly states they don’t train their AI on user data—that is unless you voluntarily submit it to the Adobe Stock Marketplace where it falls within their domain of use. Adobe’s changes to the terms of service didn’t include anything particularly incendiary in terms of AI use, but that does not mean it was entirely free from reproach.

Adobe user agreement change

Image: Adobe

Adobe User Agreement Change Also Brought up the Conversation about Privacy

The user backlash around Adobe’s terms extends beyond the fear that personal data would be used to train the company’s artificial intelligence. Adobe’s terms of service changed and, very suddenly, it brought their access to user data to the forefront of everyone’s minds. The company’s customers were faced with the realization that the software could access and use their content, even going so far as to sublicense it to others. According to 9to5Mac, the initial terms stated,

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. For example, we may sublicense our right to the Content to our service providers or to other users to allow the Services and Software to operate with others, such as enabling you to share photos.

Rewording the Adobe user agreement changes, the company later made a statement that the policy has been in place for a long time and its access to user content is restricted to the cloud. Even when cloud data is reviewed, it is done to improve their services, respond to support requests, or detect any illegal activity. “Adobe automatically scans content you upload to our services to ensure we are not hosting any child sexual abuse material (CSAM),” and the content access occurs mainly through automated systems that only review the information. A human review of private data occurs when an issue is flagged by the system or when it is requested by the user.

Adobe User Agreement Change Has Unsettled Users

The changes to Adobe’s terms of service have elicited a very strong reaction from users globally. Adobe’s extensive suite of services is an industry staple for designers and artists, many of whom sign non-disclosure agreements to maintain the secrecy around their active projects. The need to keep their content safe from prying eyes and from the potential of misuse is critical to users. 

The user backlash around Adobe’s terms isn’t the only challenge the company is facing. The Justice Department has accused the company of harming consumers by “enrolling them in its default, most lucrative subscription plan without clearly disclosing important plan terms.” The allegations go on to suggest that the company makes it very difficult to cancel a subscription and discreetly hides its cancellation charges behind fine print. Many users only find out about the fee when they go to cancel the service, further discouraging them from abandoning the tool. 

Users are growing more and more frustrated with the brand each day, but without alternative services to turn to, many are stuck in the commitments they’ve made to the subscription plans at Adobe.

tagged in Adobe AI policy change, Adobe terms of service change, Adobe terms user backlash, Adobe TOS controversy, Adobe user agreement change
Share: