a16z’s Olivia Moore on Consumer AI’s Next Chapter, the Rise of AI Agents, and Why the Market Is Still Wide Open

Consumer AI is no longer a one-player market, and it is no longer a simple story of web traffic and chatbot usage.

In a conversation with TradedVC, a16z's Olivia Moore shared a sharp read on where the consumer AI market stands now, what the latest Top 100 Gen AI Consumer Apps report reveals, and why founders still have far more room to build than many people think.

What started as an early wave dominated by ChatGPT has quickly evolved into a more layered ecosystem, shaped by mobile behavior, creative tools, voice interfaces, emerging agents, and increasingly distinct regional markets.

 

Olivia Moore, a partner at Andreessen Horowitz focused on early-stage AI investments, has become one of the most closely watched voices in consumer AI. Her firm's Top 100 Gen AI Consumer Apps report has quickly become one of the most-cited snapshots of the market, offering a recurring look at which products are gaining traction, how user behavior is changing, and where momentum is building across the category.

 

What makes the report especially valuable is that it is evolving alongside the market it tracks.

Moore explained that the project began as an internal exercise inside a16z, driven by a simple question: how are normal, non-technical users actually engaging with AI products?

 

At the time, there was no clear public benchmark for understanding what was working in consumer AI beyond anecdotes and scattered product hype. That gap created the opportunity for what has now become a widely referenced industry resource.

Over time, the report's methodology has expanded well beyond basic web traffic. What started as a website-focused ranking now includes mobile app usage and revenue indicators, reflecting a more complex and nuanced consumer AI market with each cycle. 

 

Moore noted that revenue data was important to add because traffic alone can be misleading. The most valuable users often behave very differently from casual users, and the delta between attention and monetization is only getting more important as AI businesses mature.

 

The next major piece, she said, is desktop usage.

As products like ChatGPT's desktop app, Claude, Granola, and other native AI tools become more central to daily workflows, standard web metrics are becoming less complete. Download counts may show initial interest, but they do not capture the depth or frequency of ongoing usage. For a market increasingly shaped by power users and workflow integration, that blind spot matters.

 

ChatGPT Still Leads, but the Real Story Is Market Segmentation

Even with more competition pouring into the category, ChatGPT remains the clear leader. Moore said the product still has roughly 2.5 to 2.7 times as many users as Gemini, the number-two player, across web and mobile. The gap has narrowed compared to the market's earliest days, but it remains meaningful.

Still, the more interesting trend is not just who is leading. It is how the market is splitting beneath the surface.

 

Rather than collapsing into a winner-take-all environment, consumer AI is beginning to carve out distinct ecosystems. ChatGPT, Claude, Gemini, and Perplexity may all fall under the broad banner of AI assistants, but Moore made clear that they increasingly serve different kinds of users and workflows.

 

One of the most revealing data points from the latest report is that there is only about 11% overlap between the ChatGPT and Claude app ecosystems. That is a strikingly small amount of crossover between two of the most prominent AI platforms in the market, suggesting that users are not simply bouncing between similar tools. They are choosing products for specific jobs.

ChatGPT has built strong consumer momentum across categories like travel, transportation, finance, and retail. Claude, meanwhile, has become more associated with prosumer workflows and premium data-heavy use cases. That distinction matters. It points to a market that is becoming more specialized, where interface design, connectors, product context, and workflow utility all influence which platform wins a particular user.

 

Creative tools remain another major pillar of the consumer AI landscape. Moore described creativity as the second-largest use case across both consumers and many enterprise contexts. Companies like ElevenLabs, Gamma, and Suno have remained strong across multiple editions of the report, showing a level of consistency rare in a category known for rapid churn.

 

At the same time, standalone image generators have started to lose some of their early novelty as core AI platforms absorb image creation directly into the broader chatbot experience. It is a familiar pattern in AI right now: a new category emerges, sparks excitement, gets folded into the major platforms, and then leaves room for more specialized players to rebuild the wedge at a higher level of sophistication.

 

The AI Market Is Fragmenting Globally

One of the report's most important takeaways is that consumer AI is not unfolding as one unified global market. Moore sees three major AI ecosystems taking shape: the United States and much of the broader global west, China, and Russia.

 

Censorship, sanctions, regulation, infrastructure, and domestic platform dynamics are driving that geographic splintering. In practice, it means users in different parts of the world are increasingly operating in separate AI environments with different leading products, different capabilities, and different competitive realities.

 

China stands out as the clearest example of a distinct and self-sustaining AI ecosystem. Moore pointed to players like DeepSeek and ByteDance, as well as the broader strength of Chinese AI research and product development. She also noted that Chinese companies may benefit in areas such as video and music generation because of looser copyright restrictions on training data, giving them an edge in categories where Western firms face more friction.

 

Russia has its own ecosystem as well, though it appears more mixed. Some usage flows through domestic players like Yandex's chatbot Alice, while Chinese models also meet some demand. Much of the rest of the world, by contrast, remains largely centered on U.S.-based platforms like ChatGPT and Gemini, with some regional variation in adoption of Claude and Perplexity.

 

Moore expects China's AI market to remain separate for the long term. It has the talent base, internal market size, and institutional conditions to support independent winners. Russia may eventually move closer to the western ecosystem, depending on how broader geopolitical conditions evolve, but for now, the fragmentation is real.

 

For founders, that means a global AI strategy is no longer straightforward. Distribution, model access, local competition, and even the shape of user expectations can vary significantly by region. Building for "the global AI market" increasingly means deciding which AI market you are actually entering.

 

The Highest AI Adoption Is Happening in Surprising Places

The latest report also introduced a per capita view of AI adoption, and the results were revealing.

 

Moore said the highest levels of adoption are emerging in tech-forward Asian markets such as Singapore, Hong Kong, and South Korea, as well as parts of Europe. These geographies are embracing AI tools at especially high rates, helped by white-collar labor concentration, digital fluency, and a more positive cultural posture toward new technology.

The United States, despite leading the world in foundational AI company-building, ranked only 20th in per-capita adoption.

 

That gap says a lot. The frontier of development and the frontier of adoption are not always the same. In Moore's view, cultural sentiment plays a meaningful role. Countries with more optimism toward AI and a workforce more naturally positioned to use these tools in daily work are moving faster in uptake. The U.S., by contrast, appears to have a more mixed attitude toward AI, even as its companies remain at the center of the industry.

 

Adoption was also lower across much of Africa, and lower in China and Russia within the report's comparative framing, in part because those countries operate in more distinct AI ecosystems.

 

That data suggests that the next great opportunities in consumer AI may not come only from where the models are built, but from where people are most ready to integrate AI into the fabric of everyday life and work.

 

AI Agents Are Arriving Fast, Even If the Label Does Not Last

Moore also pointed to the rapid rise of AI agents as one of the defining developments in the current market. Products capable of autonomously handling multi-step, multi-domain tasks are improving quickly, and user appetite for delegation is clearly increasing.

 

She described this shift through examples of newer agentic products that can take on more complex tasks across domains and deliver a more complete output with less manual prompting. These products represent an important leap from the earlier generation of AI tools that could generate text or answer questions but struggled to perform meaningful work end-to-end.

 

At the same time, Moore believes the term agent itself may fade sooner than many expect. Her view is that within the next couple of years, agentic behavior will become part of software. AI products will not need special labeling as agents, because users will increasingly expect software to reason, take action, and complete tasks on their behalf.

 

That future naturally favors large horizontal platforms like OpenAI and Google, which already have meaningful advantages in permissions, user context, and integrated data. But Moore also made clear that there is still plenty of room for vertical-specific agents to win. In categories where workflow depth matters more than broad platform reach, niche products still have room to build strong businesses.

 

That is especially true in markets where domain knowledge, proprietary workflow design, and user-specific context matter more than the raw breadth of a general-purpose assistant.

 

What the Best AI Founders Are Doing Differently

When asked what separates the most promising AI startups right now, Moore came back to two ideas: speed and standards.

The founders who stand out most are moving extremely fast while maintaining a very high bar for both product quality and growth. That combination has become more important because the economics and pace of software building have changed dramatically in the AI era.

Where a consumer company might once have taken years to monetize, and where a software startup reaching $1 million in annual recurring revenue once looked exceptional, AI companies are now moving much faster. Moore said many are reaching meaningful revenue or scale within a year. The market's intensity has compressed timelines and raised expectations.

Still, speed by itself is not enough. The more durable winners are not just moving fast. 

Moore pointed to ElevenLabs as a useful example. While many people assumed the biggest AI labs would quickly dominate voice, ElevenLabs has built a strong position through workflow depth, broad adoption, and a compounding data advantage that helps keep its product ahead.

 

For founders, the lesson is practical. There may be fewer obvious moats than in past software eras, but differentiation is still possible. It just tends to come from execution, workflow design, and product-specific data loops rather than from superficial novelty.

 

Voice Dictation Could Be the Next Breakout Use Case

If Moore had to pick one category likely to break out by the next edition of the report, it would be voice dictation.

She sees a voice shifting from enthusiastic behavior to a much more mainstream interface. Earlier generations of voice tools appealed mainly to users willing to experiment with new workflows or invest in specialized setups. That is changing as voice becomes more natural in both technical and non-technical work environments.

With products built on tools like Whisper and newer voice-first interfaces entering everyday office culture, more people are getting comfortable talking to their computers. What once looked niche is starting to feel normal.

The bigger opportunity, Moore suggested, is what happens when voice is paired with action. If a user speaks a request that implies a task, the system should not simply transcribe or draft the message. It should be able to carry out the next step, whether that means booking a dinner, scheduling a meeting, or running a background workflow.

That is where voice stops being just an input method and starts becoming a much more powerful operating layer for software.

 

This Market Is Still Much Earlier Than It Looks

For all the speed and noise around AI, Moore's biggest point may have been the simplest: it is still early. Very early.

 

Even ChatGPT, the most widely used AI product in the world, is still used weekly by only about 10% of the global population. And for many of those users, it remains the only AI product they regularly touch. That means the overwhelming majority of the world has either barely begun to engage with AI or has not meaningfully integrated it into daily behavior at all.

For founders and investors, that changes the picture. It means the market should not be seen as already locked up by incumbents. It is still forming. User behaviors are still emerging. Winning product categories are still being defined. Business models are still evolving. Entire geographies are still in the early innings of adoption.

Moore also pushed back on the instinct to assume that OpenAI, Google, or Anthropic will inevitably swallow every attractive opportunity. Those companies are large and powerful, but they are not limitless. They have finite engineering capacity, competing priorities, and specific strategic areas of focus. That leaves real room for independent companies to build meaningful businesses in specialized corners of the stack. Crowded does not mean closed.

 

How AI Is Already Reshaping Venture Work

Moore also gave a glimpse into how AI is starting to change venture capital itself.

She described venture diligence as a puzzle, one that involves piecing together founder instincts, customer feedback, quantitative signals, and a range of softer inputs that do not always fit neatly into a spreadsheet. That kind of work remains deeply human, but AI is already changing the supporting workflow around it.

She uses AI tools and agent-like systems to summarize meetings, organize schedules, and research founders ahead of calls. Over time, she sees much more of the repetitive side of venture becoming automatable, from sourcing and early outreach to certain forms of call analysis and first-pass pitch evaluation.

The point is not to automate judgment. It is to free up time for what matters most: spending more time with founders, building stronger relationships, and making better decisions with more context.

 

Why AI's Long-Term Impact Will Reach Far Beyond Consumer Apps

Although much of the conversation focused on consumer products, Moore's broader view of AI extends beyond rankings and app ecosystems.

She pointed to The Genome Odyssey by Ewan Ashley as an example of the kind of breakthrough AI can help accelerate, especially in medicine. In her view, one of the most exciting things about the technology is not just that it speeds up existing workflows, but that it opens the door to solving problems that previously felt too difficult, too imprecise, or simply unsolvable.

That perspective is important because it grounds the current consumer AI boom in a much larger context. The chatbots, voice products, creative tools, and agents getting attention today are only the most visible layer of a deeper transformation that will likely reshape many industries in the years ahead.

 

The Real Takeaway for Founders

For founders building in AI right now, Moore's advice is clear:

Move fast. Keep your standards high. Build with urgency. 

Find proprietary workflows or data advantages where you can. Stay focused on the real needs of a specific user rather than building something generic and hoping scale will save you.

Most of all, do not let the presence of big incumbents convince you that the market is closed. The consumer AI landscape is getting more crowded, but also more segmented. Geography matters more. Interfaces matter more. Workflow context matters more. 

Product design matters more. That gives smaller, faster, more focused teams room to win, especially in categories where specialized execution can still outpace the platforms trying to do everything.

 

The market may be moving fast, but in Moore's view, it is still in inning one. And that may be the most important signal of all.