How Consumers Are Actually Using AI // BRXND Dispatch vol 95
Dan Frommer shares data on ChatGPT's 800 million users, why it's become less about work, and the surprising generational divide in AI sentiment.
You’re getting this email as a subscriber to the BRXND Dispatch, a newsletter at the intersection of marketing and AI.
Claire here. This week, we’re spotlighting Dan Frommer’s presentation on consumer AI adoption from BRXND NYC 2025. Dan runs The New Consumer and publishes the Consumer Trends Report with Coefficient Capital. It’s a data-driven look at 5 to 10 year trends based on proprietary surveys of over 3,000 US consumers.
The main takeaway from Dan’s presentation: AI awareness is now universal. It’s higher than TikTok, Ozempic, even Doritos. And as ChatGPT has grown, it’s become less about work—73% of messages are now non-work topics, up from 53% last year. The top use cases for ChatGPT conversations are tutoring, how-to advice, and replacing Google for information searches.
To me, the most surprising finding is that millennials are now the most optimistic generation about AI, while Gen Z has become noticeably more skeptical over the past couple years. Millennials are far more excited than worried, but Gen Z is split evenly between excitement and concern. Younger consumers are still more likely to use AI—more than 60% of Gen Z used ChatGPT in June alone—but their enthusiasm has cooled. They’re more comfortable sharing personal data with AI tools and more interested in letting AI make purchase decisions, but something has shifted in how they feel about it. As a humble Gen Z-er myself, that definitely tracks—I think our perspective on AI as a broader force is shaped more by the social discourse about its impacts, even though our day-to-day use of AI sometimes feels separate from those concerns.
Watch Dan’s full talk here for all the data on how consumers are actually using AI, what they’re willing to pay for it, and the unexpected generational sentiment shifts.
What caught our eye this week
OpenAI released Company Knowledge for ChatGPT Business, Enterprise, and Edu users. It pulls context from all your connected apps—Slack, SharePoint, Google Drive, GitHub, HubSpot—and uses a version of GPT-5 trained to search across multiple sources. Every answer includes citations so you can verify the source. There are some limitations, though: it can’t search the web or create charts when this mode is on, and you have to manually enable it per conversation (for now).
Kai Williams at Understanding AI used 16 charts to explain the AI boom. There’s lots of interesting information in the article. For example: Tech giants spent $241 billion on AI infrastructure in 2024—0.82% of US GDP, rivaling the Manhattan Project and Apollo program. AI chip imports hit $200 billion annually, data center vacancy rates are at 1.6%, and demand exceeds supply. Northern Virginia dominates global data center development, with companies avoiding high-cost regions like California. The infrastructure boom is real, but the unit economics remain unclear.
WPP launched WPP Open Pro, a self-serve AI marketing platform targeting smaller enterprises who can’t afford their high-touch services. The platform uses AI agents to analyze WPP’s proprietary data, craft campaign strategies, generate creative with multiple models (Gemini, ChatGPT, Adobe Firefly), and publish directly to ad platforms.
According to a piece in Harvard Business Review, AI success depends on leadership transformation, not just technology investment. The authors argue that leaders need five critical skills to make AI work: build networks across industries and regulators to develop AI fluency, restructure processes and incentives rather than bolting AI onto legacy systems, treat AI as “a flexible teammate” in group decisions, create psychological safety for experimentation and reskilling, and personally use AI tools to model adoption.
If you have any other questions, please be in touch. As always, thanks for reading.
Noah and Claire