Eliminate Complex Prompting: Google Launches Its First Native Interactive AI Design Tool, Pics

Google officially launched a new AI-driven design and image generation application for Google Workspace at its 2026 I/O Developer Conference Pics.

This product aims to lower the barrier to professional design, allowing users to directly generate visual content such as social media images, marketing materials, and models through simple text prompts without needing complex editing skills. This move marks Google's official entry into the AI-driven design market, which is currently dominated by Canva and Claude Design from Anthropic.

To address the industry challenge of difficulty in making precise local modifications to images with generative AI, Pics introduced a bidirectional interaction mechanism. Users can not only click on specific areas to make changes using a feedback method similar to Google Docs, but they can also manually adjust specific elements like text, achieving seamless integration between generation and fine-tuning.

In terms of underlying technology, the application is powered by a new model, Nano Banana2, which focuses on enhancing precise text rendering capabilities, real-world knowledge graphs, and high-quality visual output. As a native application integrated into Workspace, Pics also supports cross-application visual collaboration, allowing users to download, print, or share.

The product has been made available to selected testers during the conference and is planned to be released this summer for Google AI Ultra subscribers. As the AI wave moves from basic multimodal content generation to deep enterprise workflow collaboration, the release of Pics not only fills the gap in Google's visual design capabilities within the office automation ecosystem, but also signals a new wave of technological opportunities in the fine-grained control and collaboration of office assets at the application layer of large models.