This week saw a flurry of activity in consumer AI, notably with the rise of Meta’s personal assistant, Muse, and its delightful mascot, Jolly. OpenAI’s Dots launched yesterday, focusing on being a personal assistant. Meanwhile, the Instinct assistant has grabbed headlines with a $10 billion valuation, thanks to its knack for booking travel, making restaurant reservations, and cancelling subscriptions.
Agentic AI has proven reliable in managing everyday tasks, prompting companies to market these services more effectively. This excitement echoes the launch of ChatGPT in 2022, making the prospect of investing in this growing field quite appealing.
However, a troubling reality lurks beneath the surface. Many of these promising technologies are encountering a barrier when it comes to consumer willingness to pay. Despite impressive technological strides, it remains uncertain whether better models will lead to a more profitable business model. This has led to a shift toward enterprise contracts and targeted vertical expansion instead of focusing solely on the consumer market.
As of May, only 2.2% of consumers were paying for AI services, averaging $31 per month. This raises an important question about the scalability of these ventures.
Andreessen notes that mainstream adoption of AI is still in its early stages. Yet, this growth appears linear, with minimal changes in both the number of customers willing to pay for AI and the amounts they are willing to spend. Even with significant enhancements in models, the jump from GPT-5.2 to Astra barely registers on the charts.
Considering these numbers, the average revenue from AI services still falls short of sustaining a business. For context, using Netflix as a benchmark, with 325 million subscribers and an average revenue of $34 per customer, that results in only $11 billion annually. This amount is less than one-third of OpenAI’s reported operating costs.
Bank of America shares a similar perspective, noting that about 3% of U.S. consumers were paying for AI in March, marking a 40% increase from the previous year. A survey by Menlo in September found that 25% of adults use AI daily, with half of those users paying for the service. Still, the overall figures highlight a systemic issue: the majority of potential users remain on the sidelines, often opting for free services.
One pressing concern regarding consumer AI is its cost. Operating AI technology is notoriously expensive, especially compared to lighter technologies like social media or cloud services. Even with a large user base, achieving profitability remains a daunting challenge.
OpenAI appears to have adapted strategically, shifting towards an enterprise-focused model that has reportedly doubled enterprise bookings since July. The launch of Dots included a strong focus on software engineers and creative agencies, illustrating how personal agents can meet business needs. Selling services to businesses at higher rates has become a viable revenue path.
However, it’s unclear how this dynamic will unfold for Muse and Instinct. Muse benefits from Meta’s expertise in personalized ad targeting, providing more monetization opportunities and time to refine its business model. Additionally, Meta is already exploring enterprise possibilities with Muse.
Instinct, on the other hand, plans to monetize by taking a percentage from purchases made through its agent. This strategy might help raise its revenue ceiling while sidestepping the heavy burden of training large models, allowing for more sustainable growth.
Despite these efforts, the harsh economic realities of consumer AI present a significant barrier to scaling without integrating enterprise revenue streams. This fundamental challenge has already been learned by many major labs in the industry and remains a constant in the ever-evolving landscape of AI technology.
