Yesterday, I addressed the controversy surrounding Adobe and its updated terms of service. Creators were outraged when they received a pop-up forcing them to accept the new terms; otherwise, they could not access Photoshop or remove the application from their computers.

It wasn’t just the mandatory nature of the terms that alarmed users. The new language suggested that Adobe could access creators’ work for various reasons. This upset many, especially those with confidentiality agreements for their work in Adobe applications. Moreover, the idea that Adobe could access work produced with their software unsettled many professionals.

Adobe remained silent until they published a blog post. In it, the company explained that the changes to their “Terms of Use” were minor adjustments aimed at clarifying their moderation policies. The company included a snippet of the terms in the blog post, highlighting the new additions and deletions in pink.

According to Adobe, the novelty is that the company now states it “may” (instead of “only”) access your content through manual and automated methods, reviewing content to detect illegal material, including child sexual abuse. If an automated system detects something suspicious, it flags it for human review. The rest of the terms remain virtually unchanged, and the pop-up was a routine reacceptance campaign for the minor changes.

Since “access” was the core of the controversy, Adobe detailed in the blog why they need it. The company explained that they need access to user content for three specific reasons: to perform standard functions in the applications (such as opening files or generating thumbnails); for cloud-based functions like Photoshop’s neural filters and the “Remove Background” tool; and to detect illegal activities or abusive content.

Adobe clarified that they do not train Firefly Gen AI models with user content and that they will never “assume ownership” of your work. However, the company uses content stored in the cloud to train its AI. Any data uploaded to Adobe’s servers can be used in this process and aggregated with other users’ data to improve Adobe’s products and services.

This is not explicitly mentioned in the blog post, but Adobe’s support article indicates that users can opt out of this AI training by going to their account privacy settings and disabling the option “Allow Adobe to analyze my content for product improvement and development purposes in Content Analysis.”

What’s the Bottom Line?

It is unlikely that Adobe is constantly reviewing your work for project secrets, and they clearly state that they will not claim ownership of your projects. However, the company can access any content uploaded to their servers: this allows them to scan for illegal content and also use your work to train their AI models.

While it is advisable to opt out of AI training, the best way to continue using Adobe applications without worrying about Adobe’s access is to keep all projects local on your machine. If you do not use Adobe’s cloud-based services, the company can only access your work for application-related tasks, such as generating thumbnails, according to the terms.

These rules have been in effect for an unspecified period: the pop-up you may have seen this week was to accept minor adjustments to the terms, not radical changes. You had already accepted these policies without knowing it. My recommendation is to limit your use of Adobe’s cloud services in the future unless absolutely necessary. The more content you can keep on your machine, the better.

Nestor Castillo, ForAllTechNews Director