A16z’s Olivia Moore says consumer AI is still early, despite gloomy economics

Consumer AI is not dead; it is still finding its real business model. That is the core argument from Andreessen Horowitz partner Olivia Moore, who says the category remains early even as revenue today is concentrated in subscriptions, technical users, and enterprise spillover.

In a new report on the top 100 consumer AI apps released Monday, Moore argued that the market is broader than the current earnings picture suggests. ChatGPT remains far and away the dominant consumer-facing AI product, but companies such as Suno and ElevenLabs appear to have durable demand, while many everyday life categories still show little to no AI competition at all.

Her view cuts against the increasingly cautious mood around consumer AI economics. The sector has attracted concern because most of the money has come from paid plans and API usage, both of which skew toward professionals and businesses rather than ordinary households. Moore says that doesn’t mean the opportunity is small — it means the market is still searching for the right way to make money from mass consumers.

Why Olivia Moore thinks consumer AI still has room to grow

Moore’s central thesis is that consumer AI should not be judged solely on how many people are paying monthly subscription fees. In her view, the more interesting question is whether companies can create products that feel free or low-friction to use while monetizing through models more familiar to the consumer internet, including advertising and optional upgrades.

That position matters because the current state of the category looks lopsided. The biggest revenue drivers so far are people using AI for coding, automation, and other technical tasks, which often require stronger models and higher-value paid plans. That has made the economics look closer to enterprise software than to classic consumer internet products.

Moore acknowledged that reality, but said it does not necessarily define the future. The reason, she suggested, is that consumer AI companies may not need frontier models for every task. Lower-cost systems, including open-source options, could be good enough for a growing share of everyday use cases.

How does consumer AI make money right now?

Right now, consumer AI mainly makes money from subscriptions and usage-based fees. Moore said that is one reason the category feels financially narrow: paying customers tend to be professionals, power users, or teams that need high-capability tools for work.

That revenue structure has created a perception problem. If only a small slice of households is paying for AI, the market can look tiny. But Moore argues that this may say more about the current monetization model than about total demand.

Moore’s view is that the consumer market may be healthier if users are not forced to pay directly out of pocket for every interaction, and instead can access products for free or cheaply while companies monetize in more familiar consumer ways.

That logic would bring consumer AI closer to the broader internet playbook. Search, social media, and many media products became massive businesses not because every user paid a subscription, but because a large audience could be monetized indirectly.

What the top 100 consumer AI apps reveal

Moore’s report on the top consumer AI apps highlights both concentration and white space. A few products are already proving they can keep users coming back, while several large consumer categories remain largely untouched.

At the top of the market, ChatGPT still dominates. But the list also shows signs that some specialized apps have staying power, especially in audio and music generation. The fact that companies like Suno and ElevenLabs continue to stand out suggests that consumers are willing to adopt AI tools when they solve a clear, repeatable problem.

At the same time, the report underscores how incomplete the market still is. Moore pointed to several areas that appear to have almost no meaningful AI presence yet, including social apps, dating, marketplaces, retail, travel, finance, and health.

That gap is important because those are the kinds of categories where consumer behavior is already frequent, habit-driven, and commercially valuable. If AI can break into them, the market could expand well beyond the current crop of productivity and creation tools.

How are AI companies lowering costs?

AI companies are trying several approaches to reduce the high cost of serving consumers, and Moore says the trend is moving in the right direction. One obvious lever is product tiering. OpenAI’s lower-priced Go plan, which costs $8 per month, appears designed to reach a broader audience than its premium offerings.

Another strategy is using cheaper models for tasks that do not require the most advanced systems. Moore argued that many everyday consumer needs do not call for frontier intelligence. A lightweight model can often handle summarization, simple writing, planning, or routine assistance at a much lower cost.

Open-source models may also play a larger role. According to Moore, founders are increasingly talking about building consumer products on top of open-source infrastructure, which could lower costs and make it easier to create services with more sustainable margins.

That shift matters because consumer AI has a different economics problem from traditional software. In classic cloud products, incremental usage is relatively cheap. In AI, every interaction can create a material cost, which makes pricing and retention far more difficult.

Why technical automation changed the revenue picture

Technical automation has become one of the clearest monetization engines in consumer AI, but it also distorts how the category is understood. The people generating the strongest revenue are often not casual consumers; they are builders, coders, and advanced users who need high-end capabilities.

That dynamic helps explain why consumer AI companies can look successful on revenue charts while still serving a narrow user base. Moore suggested that this is a transition stage rather than the endpoint of the market.

As more products are built for non-technical users, the model itself may become less central than the outcome it produces. In that scenario, consumer AI could look less like a set of coding tools and more like a general-purpose layer embedded across everyday apps and services.

Are consumer and enterprise AI still separate markets?

Moore’s answer is increasingly no. She argued that AI has blurred the old line between consumer and enterprise products, creating a category she described as effectively “prosumer” in nature.

Before the AI boom, consumer-first products often spent years building a user base before adding team plans or business features. AI companies are moving much faster. Some start with individuals and become enterprise-relevant within months.

She pointed to examples such as Gamma, ElevenLabs, and Cursor, which began as tools with consumer appeal but quickly became more valuable to teams and businesses. In that sense, the consumer label can be misleading: many AI apps that look consumer-facing on the surface are actually designed for paid professional workflows from the beginning.

That speed of expansion is one reason investors remain interested, even with weak near-term consumer monetization. The real prize may not be a pure consumer app at all, but a product that can travel from individual adoption to company-wide usage quickly.

What counts as true consumer AI?

Moore believes the industry has not yet seen the full version of consumer AI. In her view, many of the most visible apps are still tools for creators, builders, or workers, not the kind of everyday products that can reach broad mainstream audiences.

That distinction matters. A tool used by developers to write code is still consumer-facing if an individual signs up for it, but it does not tell us much about how ordinary users will adopt AI in shopping, entertainment, relationships, travel, or health.

Moore said the absence of strong entrants in those areas is exactly what makes the market feel unfinished. The next phase, she argued, depends on AI showing up where people already spend time and money in daily life.

• Current leaders are mostly productivity and creation tools.
• Mainstream consumer categories remain largely open.
• Future winners may rely on ads, freemium models, or other indirect monetization.
• Lower-cost models could expand adoption by reducing serving costs.
• Many “consumer” AI apps are already drifting toward enterprise use.

Why the economics look bleak — and why investors still care

The skepticism around consumer AI comes from a real place. If only a small share of households is paying, and if each interaction is expensive to serve, the path to profitability can look uncertain. That has led some observers to argue that the current wave of consumer AI is overhyped.

Moore did not dismiss those concerns. Instead, she reframed them. The current model may be narrow because the industry has not yet figured out the mass-market version of AI. In other words, the problem may be less about demand and more about fit.

For investors, that is what keeps the category alive. A market that is economically awkward but structurally early can still become enormous if the right use cases emerge. The fact that major categories like travel, retail, health, and social have not yet produced clear AI leaders suggests that the story is far from over.

That helps explain why consumer AI remains one of the most watched parts of the startup ecosystem. Even when the revenue is concentrated and the market looks uneven, the upside of finding a product that works for millions of ordinary users remains too large to ignore.

What happens next for consumer AI?

The next six months may be decisive for whether consumer AI stays a niche powered by power users or expands into a much broader business. Moore thinks the industry needs to see meaningful movement in categories that touch everyday life, not just work and creation.

If that happens, the economics could shift quickly. Products built around ads, free access, and lightweight usage could attract far larger audiences than subscription-only tools. At the same time, cheaper models and open-source infrastructure may make it easier for founders to build at sustainable margins.

For now, though, Moore’s report sends a clear message: consumer AI is not mature, and the market’s biggest opportunities may still be ahead. ChatGPT may be the headline act, but the deeper story is how much of consumer behavior has yet to be reshaped by AI at all.

For now, Moore’s bottom line is cautious but optimistic: consumer AI may look incomplete, but that incompleteness is exactly why it still matters.