a16z report finds ChatGPT and Gemini achieve over 50% retention, outpacing all AI rivals

The a16z “100 Gen‑AI Apps” study, released this week, shows that OpenAI’s ChatGPT and Google’s Gemini each retain more than half of their users after the first month—a metric that eclipses every other generative‑AI product surveyed (a16z, 2025). Retention, a leading indicator of long‑term engagement and monetization potential, was calculated by tracking active accounts across a cross‑section of consumer‑facing AI tools. The report notes that while many niche chatbots and image generators see churn rates exceeding 70%, ChatGPT and Gemini consistently hold users at a 52‑55% level, suggesting a durable habit loop driven by conversational depth, multimodal capabilities, and integration into daily workflows.

Japanese tech outlet ZDNet corroborated the findings, highlighting ChatGPT’s position at the top of the market and Gemini’s rapid ascent to second place (ZDNet Japan, 2025). The article attributes Gemini’s rise to Google’s aggressive rollout of new features, such as real‑time translation and code assistance, which appear to resonate with both developers and enterprise users. ZDNet’s analysis also points to the breadth of the two platforms’ ecosystems—ChatGPT’s plugin architecture and Gemini’s seamless tie‑ins with Google Workspace—as key factors that keep users returning week after week.

The retention advantage translates into a competitive moat in a crowded field where dozens of startups vie for attention. According to a16z, the high‑engagement cohort of users not only drives higher subscription revenues for OpenAI and Google but also creates a feedback loop that accelerates model improvement through richer interaction data. In contrast, lower‑retention apps struggle to justify the heavy compute costs associated with large‑scale inference, limiting their ability to scale pricing or invest in next‑generation research.

Industry observers see the data as a bellwether for the broader AI market’s maturation. The a16z report emphasizes that “sticky” usage patterns are emerging as the primary differentiator among AI offerings, shifting the competitive focus from raw model size to product integration and user experience. As ZDNet notes, the Japanese market’s appetite for AI‑enhanced productivity tools underscores a global trend: enterprises are prioritizing platforms that can be embedded into existing workflows rather than standalone novelty apps. This alignment of user retention with business utility suggests that ChatGPT and Gemini are likely to capture a disproportionate share of future AI spending.

Finally, the findings raise questions about the sustainability of the current leader‑board. While ChatGPT and Gemini enjoy a clear retention edge today, the report warns that emerging rivals with specialized vertical solutions could erode the gap if they achieve comparable stickiness. Nonetheless, the a16z data provides a quantitative benchmark: any AI product aspiring to compete at scale must now aim for the 50%‑plus retention threshold that ChatGPT and Gemini have set (a16z, 2025).