A16z's Olivia Moore Outlines the Economics and Untapped Sectors in Consumer AI
Andreessen Horowitz partner Olivia Moore discusses why consumer AI is still in its infancy, from prosumer revenue dominance to major untouched categories.
The consumer artificial intelligence landscape is frequently characterized by debate over unit economics and monetization models. In a newly released report ranking the top 100 consumer AI applications, Andreessen Horowitz partner Olivia Moore provided an assessment of where user interest and spending currently lie. While OpenAI's ChatGPT continues to maintain a substantial lead, smaller platforms such as Suno and ElevenLabs are demonstrating consistent market retention, pointing toward an evolving consumer ecosystem.
Despite the prevailing skepticism around consumer software economics, Moore suggests that the sector is still in its initial phases. A closer look at the data shows that much of what currently registers as consumer adoption is tied to professional workflows, while traditional consumer verticals remain largely unaddressed.
The prosumer reality behind current AI revenue
While venture-backed services are often labeled as consumer applications, user spend remains heavily anchored in technical and professional automation. Moore pointed out that pre-AI companies like Canva historically spent six or seven years operating as consumer-first tools before rolling out enterprise or team tiers. Today, platforms such as Gamma, ElevenLabs, and Cursor frequently transition to majority-enterprise revenue within 18 months.
According to Moore, consumer AI spending is currently concentrated across three main categories:
- Product-building tools: Platforms like Lovable, Replit, and Fal, which assist users with technical creation.
- Product marketing: Video and ad generation tools, including Higgsfield and HeyGen.
- Work management: Workflow automation software such as Manus, Fireflies AI, and Granola.
Because these products cater primarily to high-utility or revenue-generating tasks, users are more willing to fund them individually before the software expands into team-level corporate procurement.
Rethinking monetization and inference costs
Direct consumer monetization remains constrained under current software models. Citing the State of Markets report, Moore noted that only 2.2% of U.S. households currently pay for AI services out of pocket. In Silicon Valley, high-income tech workers with corporate budgets often default to subscriptions, but broader audiences historically prefer ad-supported, free access with optional paid tiers to remove advertisements.
Compounding this challenge is the marginal cost of serving AI queries, which remains higher than delivering traditional web services like Google Search or Facebook. To expand margins, developers are beginning to tailor their infrastructure to task requirements. Moore highlighted OpenAI's $8 per month ChatGPT Go plan as an example of cost segmentation, noting that not every consumer task demands a flagship frontier model. For applications where technical automation is not the core focus, lower-cost and open-source models are increasingly viable.
Whitespace across traditional consumer verticals
One of the most notable takeaways from the top 100 list is the absence of generative AI applications in core consumer categories. Moore noted that sectors such as social apps, dating platforms, marketplaces, retail, travel, personal finance, and healthcare currently have no representation among the top 100 consumer AI apps.
Because developers have prioritized technical tasks and code generation—areas where frontier model reasoning is essential—everyday consumer software remains virtually untouched by dedicated generative AI entrants. Moore indicated that filling these category gaps will be critical over the next six months to realize true consumer AI adoption.
What it means for developers
For engineering teams building AI applications, the transition from prosumer tools to mainstream consumer apps highlights key architectural and operational priorities:
- Matching model scale to user requirements: High-cost frontier intelligence is necessary for coding and technical automation, but mainstream consumer features—such as recommendation feeds, basic drafting, or conversational interfaces—often perform well on lightweight alternatives. Deploying lower-cost models helps keep unit economics viable.
- Flexible API testing: Developers testing combinations of specialized and general models can evaluate options through platforms like Apixo, which offers token-based access to leading AI models via a single API key to simplify cost benchmarking.
- Exploring alternative business models: With just 2.2% of households paying subscription fees for AI, relying solely on recurring monthly memberships may limit total addressable market size. Hybrid models, including ad-supported tiers or metered usage, will likely be necessary for broad consumer adoption.
- Targeting greenfield verticals: Software categories outside of developer tooling and marketing generation—such as healthcare, travel, and personal finance—present large openings for teams aiming to build consumer-facing applications before incumbents establish dominance.
Source: A16z’s Olivia Moore on the state of consumer AI — TechCrunch AI. Written by the Apixo team from that report.
One key for Claude, GPT, GLM, DeepSeek and more. Pay per token with crypto.
Get your API key
