Google Search adds free AI image generator in US, India and Japan

Google Search’s Search Generative Experience (SGE) now produces four AI‑generated images from a text prompt at no cost, but the feature is limited to users in the United States, India and Japan who have opted into Search Labs [1][2]. This gives content creators a free alternative to paid tools like DALL‑E or Midjourney, while Google tests safeguards such as watermarks and metadata labeling.

How the tool works

Users who have joined the SGE testing program can type a prompt—e.g., “draw a chicken in the style of Van Gogh”—directly into the Google search bar. Within seconds the interface displays four generated images, each with an “Edit” button that lets the user refine the prompt for more detail [2][4]. The images carry a visible watermark and are tagged with identifying metadata to signal they were AI‑created [4]. Google also plans an “About This Image” description to provide context, though that feature is still under development [4].

Market context and competition

Google’s move mirrors Microsoft’s recent rollout of DALL‑E 3 in Bing Chat, but unlike Bing’s separate Image Creator, Google embeds generation directly into the search results page [2]. By leveraging its proprietary Imagen model rather than OpenAI’s DALL‑E, Google hopes to differentiate quality and style handling, offering users a seamless, no‑subscription experience [2]. The free tier contrasts with the paid subscriptions required for comparable outputs from Midjourney or OpenAI, potentially expanding Google’s share of the generative‑AI image market.

Restrictions and rollout

The feature is currently restricted to adult users (18+) in the three regions and only accessible through Chrome or the Google app on iOS and Android [4][2]. Google’s generative‑AI policy blocks harmful, misleading, or explicit content, and every output is watermarked to curb misinformation [4]. While the tool is still experimental, Google hints at broader integration, possibly extending the generator to Google Images and other products in the future [4][5].

What to watch

• Geographic expansion: Whether Google opens the AI image generator to additional countries beyond the US, India, and Japan.
• Model updates: Introduction of newer models (e.g., Gemini 3) that could replace or enhance the current Imagen‑based output.
• Competitive response: How Microsoft and OpenAI adjust their own search‑integrated image tools in reaction to Google’s free offering.

Google’s free, search‑embedded image generator signals a push to make generative AI a default web‑search capability, testing user adoption and policy enforcement before a wider rollout. The outcome will shape how search engines compete for the growing demand for on‑the‑fly visual content.