Adobe Teams up with Google Gemini – Here’s Why That’s Complicated

Adobe announced at Google I/O this week that it is bringing its Adobe for creativity connector to Google Gemini. Once live, Gemini users will be able to describe what they want to create and have Adobe’s tools handle the execution. You’ll be able to generate images, resize assets, produce video variations: all without leaving the Gemini interface.

It’s a logical expansion of a product that already exists for Anthropic’s Gemini competitor, Claude. It also comes with a few things worth thinking carefully about, which we’ll get to.

The connector already exists for Claude, where it gives you access to 50+ pro-grade Adobe tools. The Gemini version will work the same way. Adobe says hundreds of millions of Gemini users will gain access to its imaging, design, and video capabilities “in the coming weeks.”

What Adobe Says

Adobe’s framing is outcome-driven creation: you describe what you want, the agent figures out how to get there. In a press release, the company highlights two user stories: a photographer who got cinematic results without jumping between Photoshop and Lightroom, and a social media creator who produced every platform format he needed from a single photo inside Claude.

The workflow angle is genuinely useful. Fewer context switches, faster iteration, approval checkpoints before the agent moves forward. These are real benefits for working creators. The Gemini expansion continues that logic. Adobe’s stated vision is that its pro tools should be available “wherever creative work happens, not just inside our own apps.” That part, at least, is consistent.

However…

On paper, it’s a straightforward expansion. In practice, it raises a few questions worth considering.

First and foremost, I’m not disputing the usefulness of certain AI editing tools. It sure makes editing faster and more efficient, and it can be a lifesaver if you prefer being in the field with your camera to sitting at a computer for hours. However, blurring the line between AI and photography further is not something I’m too excited about.

On the plus side (sort of), Firefly was trained on licensed content: Adobe Stock, public domain material, and content Adobe had explicit rights to use. That was a meaningful distinction in a market full of AI models trained on scraped internet data without permission.

However, Google’s AI doesn’t share that background. The company has faced significant criticism over its training data practices. Partnering with Google doesn’t make Adobe’s own model less clean, but it does cast a shadow over it. If the ethical sourcing argument is central to why creators should trust Adobe’s AI ecosystem, teaming up with one of the more controversial players in the space is worth noting.

Whether the tradeoffs are worth it depends on how much you trusted Adobe’s ethical AI story in the first place, and how you feel about Google being in the room. If you’re eager to test it out, I don’t judge, but I am saying – be mindful about it. The creativity connector is coming to Gemini soon, and you should be able to test it out over the coming weeks.

---