According to ZDNet, a recent Artificial Analysis survey of 300 developers and creators during Q3 2025 revealed Google Gemini leads AI image creation with 74% adoption, followed by OpenAI’s GPT Image (64%) and Black Forest Labs’ Flux (37%). The survey, conducted before OpenAI’s Sora 2 and Video API releases, showed Gemini dominates both personal (77%) and enterprise (68%) use for images, while Google’s Veo leads video creation for personal (67%) and enterprise (72%) use. The findings align with Similarweb data showing Gemini gaining on ChatGPT, with users citing integration with Google Docs and Gmail as key advantages. This growing preference for Google’s AI tools signals a significant shift in the competitive landscape that warrants deeper analysis.
Table of Contents
The Integration Edge in Enterprise AI
Google’s dominance in this survey isn’t surprising when you consider the strategic advantage of their ecosystem integration. Unlike standalone AI tools, Gemini benefits from being embedded within the Google Workspace environment that millions of businesses already use daily. When artificial intelligence tools are seamlessly integrated into existing workflows, adoption barriers drop significantly. Enterprise users aren’t just choosing the best AI model—they’re choosing the path of least resistance that minimizes workflow disruption. This explains why respondents highlighted the “fluid” transition from research in Gmail to content creation in Gemini as a decisive factor.
Broader Market Implications and Competitive Response
The survey results should alarm competitors, particularly Microsoft and OpenAI, despite their early market leadership in other AI segments. While ChatGPT captured the public imagination with conversational AI, Google is winning the practical application battle where it matters most—in daily productivity tools. The timing is particularly concerning for rivals, as the survey predates Google’s late-August release of their popular “nano banana” image model for free within Gemini, which likely accelerated adoption further. We’re witnessing a classic platform play where Google’s extensive ecosystem creates a moat that pure-play AI companies struggle to cross.
Understanding Enterprise Adoption Patterns
The gap between personal (77%) and enterprise (68%) adoption of Gemini for image creation reveals important insights about organizational AI integration. Enterprises move slower due to compliance, security, and workflow considerations, yet 68% adoption in such a short timeframe is remarkable. The fact that 44% of organizations already use AI-generated images in production workflows indicates we’ve moved beyond experimentation into practical implementation. However, the preference for AI video in advertising (55% vs 42% for images) suggests companies are being strategic—using video where it delivers the highest marketing ROI while cautiously expanding to other use cases.
Why Quality Trumps Cost in Enterprise Decisions
Perhaps the most telling finding is that enterprises prioritize “higher-quality outputs” (82%) and “accurate prompt adherence” (73%) over “lower total cost” (55%) when selecting AI tools. This reflects a maturity in enterprise AI strategy—companies recognize that the real cost isn’t the subscription fee but the time wasted on regenerating or editing poor outputs. This quality-first approach plays directly to Google’s strengths, given their extensive research in AI model training and multimodal understanding. As organizations scale AI usage, reliability becomes more valuable than novelty.
The Coming Competitive Battle
While Google currently leads in creative AI applications, the landscape remains fluid. OpenAI’s forthcoming Sora 2 and Video API releases could shift the balance, particularly if they achieve superior output quality. However, Google’s integration advantage will be difficult to overcome unless competitors develop equally seamless workplace integrations. The real battle may unfold in the specialized enterprise space, where companies like Adobe are integrating AI directly into creative tools like Photoshop. The winners will likely be those who understand that in enterprise AI, context and workflow integration matter as much as raw model capability.
