Consumer AI Has a Revenue Problem, a16z Finds
Consumer AI has its business model backwards, said Olivia Moore, a partner at Andreessen Horowitz covering consumer AI, with the release of the seventh edition of the firm's Top 100 Gen AI Consumer Apps report. TechCrunch
The report was posted on October 5, 2026, after three years tracking consumer AI. It ranks the top web and mobile AI products by monthly traffic. a16z The method has stayed consistent while the names on the list have turned over.
ChatGPT is the biggest player in the top 100 report. Suno and ElevenLabs were cited as showing staying power among consumer AI apps. The report also details half a dozen consumer categories untouched by AI.
Turnover has been fast. New entrants often spike on launch traffic and then fade. Holding a position across editions is the exception, because retention is harder than initial reach.
Monetization is the central tension. Moore said almost all AI revenue so far has come from subscriptions and token usage. Subscriptions are flat monthly fees. Token usage is a metered charge based on how much text or media the model processes. She said consumer AI needs revenue streams beyond subscriptions and API charges, the fees developers pay to plug a model into their own product. The State of Markets report found that 2.2% of U.S. households are paying for AI.
Pricing pressure is visible at both ends. ChatGPT offers a Go plan costing $8 per month. Google AI offers a $20-per-month Pro subscription and a $249-per-month Ultra subscription. a16z The range covers casual users, power users and professionals inside the same subscription container.
The enterprise side shows a parallel pattern. Seema Amble is a partner at Andreessen Horowitz. Moore and Amble reported that companies are still adopting a range of different AI products for certain tasks. That finding was the subject of an October 2, 2025 article about which AI companies startups are actually paying for. TechCrunch
The series has its own lineage. The sixth edition credits Justine Moore, Anish Acharya, Olivia Moore, Bryan Kim, and Zach Cohen. a16z It includes a chart titled "ChatGPT has the Lead in Consumer Subscriptions". Moore discussed the latest edition on an a16z podcast. a16z Earlier, Moore was an investor at venture capital firm CRV alongside Justine Moore. TechCrunch
The broader context for builders is that the constraint is not distribution. Monthly traffic can be acquired. Paid conversion is the bottleneck. A 2.2% household penetration rate for paid AI leaves little room for a market where nearly every product charges the same way for general access.
In my view, the untouched categories are as instructive as the leaders. Where AI has not yet embedded in a consumer workflow, the opening is not for another chatbot with a subscription attached. It is for products where model cost is absorbed into a transaction, marketplace fee or outcome the user already understands. That shift will be awkward. It is also how consumer software has historically broken out of early adopter pricing.
Worth flagging for technical readers is that token usage and API charges map cleanly to developer costs. They map poorly to consumer value. Subscriptions solved that mismatch for a time by offering predictability. The next phase will likely decouple what the user pays for from how the model is metered underneath.
