SAN JOSE, CA – Adobe is revising the terms customers must agree to when using its applications in an effort to restore trust and clarify its stance on AI training. The change, announced in a blog post, follows a week of backlash from users who feared an update to Adobe’s terms of service would permit the use of their work for AI training. Critics also expressed concern about Adobe’s content moderation policies, which could allow Adobe employees to review customer work under certain circumstances.

Following its prior Terms update, many Adobe customers took to social media to denounce the company’s approach, which appeared to grant Adobe a license to use any and all content created by its customers for practically any and all purposes.

I just cancelled my Adobe licence after many years as a customer.

The new terms give Adobe “worldwide royalty-free licence to reproduce, display, distribute” or do whatever they want with any content I produce using their software.

This is beyond insane. No creator in their… pic.twitter.com/8UK3ur3WtH

— Sasha Yanshin (@sashayanshin) June 7, 2024

The updated terms of service, which are set to roll out on June 18, aim to provide clearer guidelines on what Adobe is allowed to do with its customers’ content, according to David Wadhwani, Adobe’s president of digital media.

“We have never trained generative AI on our customers’ content, we have never taken ownership of a customer’s work, and we have never allowed access to customer content beyond what’s legally required,” Wadhwani told The Verge.

Adobe faced significant scrutiny from creatives over the past week after its customers were alerted to vague language in its terms of service update regarding AI. Customers interpreted this language to mean the company was granting itself permission to access and use customers’ work to train its generative AI models. Although Adobe’s policies around AI training were unchanged, Scott Belsky, Adobe’s chief product officer, acknowledged that the wording was “unclear” and emphasized the importance of “trust and transparency.”

“In retrospect, we should have modernized and clarified the terms of service sooner,” Wadhwani said. “We should have proactively narrowed the terms to match what we actually do and better explained our legal requirements.”

Despite the sweet talking and assurances from Adobe, the company’s stock price has taken a huge hit since the controversy erupted online.

like watching an abusive relationship slowly crash and burn https://t.co/0NX1Z5pHbY pic.twitter.com/Po8SaTfF7Y

— M. A. Zavala 💖💜💙 (@MegZavala) June 10, 2024

Additional Concerns from Creators

Beyond concerns about AI training, Adobe’s loose, wide-reaching language in the terms of service update has irked many creators. The vague terms could potentially allow Adobe to scan, view, and review any content passing through Adobe apps or Adobe Cloud servers. This has raised alarms among professionals who use Adobe products for sensitive work.

One significant example is content protected by nondisclosure agreements (NDAs). Creators who sign NDAs must keep such files confidential until the agreement expires. Understandably, they are wary of Adobe potentially accessing content they are not authorized to disclose.

Moreover, some creatives have encountered resistance when attempting to address these issues with Adobe. Earlier this week, conceptual artist Sam Santala posted on X about his frustration with not being able to speak to an Adobe customer service representative, cancel his subscription, or even uninstall Photoshop without first agreeing to the new terms of service.

Adobe maintains that it only scans files stored on its cloud service, not on users’ personal devices. “Adobe performs content analysis only on content processed or stored on Adobe’s servers; we don’t analyze content processed or stored locally on your device,” the company stated. This verbiage remains unchanged.

Is Adobe Too Dominant?

The creative community has long expressed concerns about Adobe’s industry dominance, subscription-based pricing models, and use of generative AI. While Adobe trained its Firefly AI model on Adobe Stock images, openly licensed content, and public domain content to address ethical concerns, some artists have discovered images referencing their work on Adobe’s stock platform, raising questions about the effectiveness of the company’s protections.

Wadhwani expressed confidence in Adobe’s content moderation process but acknowledged it is “never going to be perfect.” He stated that Adobe can remove content that violates its policies from Firefly’s training data and that customers can opt out of automated systems designed to improve the company’s services.

In its blog post, Adobe emphasized that “trust must be earned” and committed to taking feedback into account to inform the new changes. While greater transparency is welcomed, convincing disillusioned creatives of Adobe’s good intentions may take time. “We are determined to be a trusted partner for creators in the era ahead. We will work tirelessly to make it so,” Wadhwani said.