ChatGPT launched as a “research preview” in late 2022. Less than three years later, in July 2025, it has shattered all records for technology adoption. The platform now boasts over 700 million weekly active users, a staggering figure that represents nearly 10% of the world's adult population. The sheer volume of use is breathtaking, with users sending more than 2.5 billion messages per day, or about 29,000 messages per second. But what's truly captivating is what people are doing with all that power.

While many expected this powerful tool to be a workplace revolution, the data reveals a different story. The majority of consumer use is, in fact, not for work. As of July 2025, over 70% of consumer-level queries were unrelated to employment, a share that has grown dramatically from 53% just a year prior.
This suggests that the economic value of generative AI may extend far beyond professional productivity, having a massive impact on personal lives and “home production”. The study also hints at a staggering consumer surplus, with one paper estimating that U.S. users would need to be paid $98 to give up using generative AI for a month, implying a market value of at least $97 billion annually.
Forget what you've heard about AI being a coder's tool or a digital therapist. The data shows that the primary use cases are surprisingly mundane, yet immensely powerful. The three most common topics— “Practical Guidance,” “Seeking Information,” and “Writing” —account for a combined 77% of all conversations.

Seeking Information is Exploding: While “Writing” and “Practical Guidance” are popular, “Seeking Information” is the fastest-growing category, increasing from 14% to 24% of all usage between July 2024 and July 2025. This highlights ChatGPT's growing role as a powerful alternative to traditional search engines, providing customised, conversational answers.
The Writing Revolution Isn't What You Think: “Writing” is the leading use case for work-related tasks. However, nearly two-thirds of these messages are not requests to write from scratch, but rather to edit, critique, or summarise existing text. This shows that professionals are using AI not as a replacement, but as an indispensable co-pilot for refining their own work.
A Niche for Nurture: Contrary to popular belief, “Computer Programming” and “Relationships and Personal Reflection” make up a surprisingly small portion of messages, at just 4.2% and 1.9% respectively.

The typical ChatGPT user is rapidly changing.
The Gender Gap is Closing: The early user base was disproportionately male, with about 80% of active users having typically masculine names in the first few months. However, that gap has narrowed dramatically, reaching near-parity by mid-2025, and now with a slight lean towards users with typically feminine names.
A Young and Global Audience: Nearly half of all messages (46%) sent by adults come from users under the age of 26. Furthermore, the fastest adoption growth is happening in low- and middle-income countries, suggesting a democratising effect for this technology.
Educated Users Lead the Way: Users with a bachelor's degree or higher are significantly more likely to use ChatGPT for work. This trend highlights the technology's immediate impact on knowledge-intensive jobs.
Search & Discovery Disruption: With 24% of usage being “Seeking Information,” ChatGPT is becoming a significant search alternative. Traditional SEO and paid search strategies need adaptation.
Content Marketing Evolution: The dominance of “Practical Guidance” requests suggests consumers want personalised, actionable advice rather than generic content. Brands need more consultative approaches.
Customer Service Transformation: Users clearly prefer AI for information gathering and decision support, indicating potential for AI-powered customer interactions.
Product Research & Recommendations: The 24% seeking information + practical guidance suggests users want purchasing advice. Affiliates could create AI-powered recommendation tools.
Educational Content: With 10.2% of messages being tutoring/teaching requests, there's opportunity for educational affiliate content in various niches.
Writing & Content Tools: Given the prominence of writing assistance (editing, critiquing, creating), affiliates could promote AI writing tools, courses, or services.
Demographics-Specific Targeting:
Localisation Opportunities: Rapid growth in lower/middle-income countries suggests untapped affiliate markets with less competition.
Source Attribution: The paper shows ChatGPT users frequently seek “Seeking Information” (24% of usage) and “Practical Guidance” (29%). When LLMs provide this information, they increasingly cite sources, making citation a new traffic channel.
Authority Building: Being referenced by AI systems signals credibility and expertise to users who may never have discovered your content through traditional search.
Scale Amplification: With 2.5+ billion daily ChatGPT interactions, a single piece of well-cited content could reach exponentially more people than traditional organic reach.
Beyond Traditional SEO: While Google optimization remains important, content creators need to optimise for LLM training data and real-time retrieval systems. This means:
Quality Over Quantity: LLMs tend to cite authoritative, well-researched content. The spray-and-pray content approach becomes less effective than creating definitive resources.
Real-Time Relevance: Unlike traditional SEO which can take months, LLM citation can happen immediately as systems access current web content for responses.
Become the Definitive Source: Create the most comprehensive, accurate resource on specific topics in your niche. LLMs gravitate toward authoritative, complete information.
Structured Knowledge: Present information in clear, factual formats that LLMs can easily parse and cite – think data tables, step-by-step guides, and well-organised reference materials.
Citation Tracking: Monitor when your content gets referenced by various AI systems to understand what type of content gains traction.
Multi-Platform Strategy: Don't just optimise for one LLM – different systems may have different citation preferences and training data sources.
The research suggests we're moving toward an information ecosystem where being cited by AI systems becomes as valuable as ranking #1 on Google, potentially more so given the personalised, conversational nature of AI interactions.