Inside the big promise Adobe is making to advertisers with its generative AI tool

Its big pitch is that its huge stock of images – on which it can train AI tools for generative purposes – is allowing it to circumvent one of the largest ongoing issues with GenAI. As arguments boil over as to the use of online data to train up AI, and who should be compensated for any profit made from those AI tools, Adobe does not have that issue.

In fact, so confident is it that it owns the images on which the AI is trained that it is promising to indemnify its users from any disputes that arise as a result of the tool. It is an extension of Adobe’s existing indemnification rules – but repackaged in a way that solves one of the most fundamental early issues with AI for marketers.

​

Given the difficulties in demonstrating exactly which sources were used to create generative AI pieces, this also has the benefit to Adobe of keeping its users within its own programs and tools where demonstrating provenance is easier.

Shantanu Narayen is CEO of Adobe Systems. He explains: “We believe that few companies can create the foundational models Adobe can with Firefly. This enables us to build these high-quality foundational models that are safe for commercial use. Our AI approach is built with transparency at the center… all content in Firefly will automatically be tagged with content credentials.”

David Wadhwani is Adobe’s president, digital media. He explains that the use of the GenAI tools deployed by Adobe across its suite of products since the launch of Firefly has outstripped its expectations: “Two major pain points… the first is that demand for content has grown so fast that creative teams are having trouble keeping up. The second is that marketing is now happening at the speed of social. Marketing teams need to be much more agile to compete with all the noise in the market today.”

At the Adobe Summit EMEA it was very clear that, like most other tech companies, Adobe is keen to tout its AI bonafides. Its own research makes plain the extent to which safety is part of its core offering. It states that 34% of customers are prioritizing actions like having guardrails in place to encourage ethical use, protect creators’ intellectual property and avoid litigation.

It states that only 10% said companies would not use generative AI at all.

Training wheels

The news follows suggestions from the Japanese government that it too will be making changes to prevent any copyright issues with the images and data used for training AIs. Notably, while this would not prevent any copyright issues with the actual output from said generative AI it would allow the country to keep up with wider developments in artificial intelligence development. Training data, then, has quickly become a flashpoint for GenAI.

Other key topics at the Adobe Summit, explained in the opening keynote by a series of senior Adobe team members, were the use of data for personalization and the ability to do contiguous customer journeys online. Artificial intelligence was threaded through every announcement, from Adobe Express’ integration of GenAI to the insight-led approach of its cloud businesses.

​

Narayen explained: “Creativity is the new productivity. Generative AI also represents this incredible new technology frontier… this is going to revolutionize technology as we know it. It has the ability to become this incredible co-pilot. For the world’s creative minds, I think it rapidly expands access to creativity. For businesses, it represents an opportunity for automation… and personalization.”

For marketers looking to go deep into the use of generative AI for creative, that indemnification and provenance through the use of Adobe’s own tools could be too powerful a lure to ignore.