The true value proposition of LLMs lies in their ability to convert unstructured data from sources like websites and documents into structured information with reasonably high accuracy. Yet, the real profit lies in the products built on top of LLM technology.
Each year, approximately 4 million books are published worldwide. On average, a book contains fewer than 120,000 words, translating to less than 160,000 tokens in LLM (Large Language Model) terms. Imagine if every single one of these books were generated by GPT-4—it would amount to an astounding 640 billion tokens. At $5 per million tokens, generating all these books would tally up to about $3.2 million! Let’s say the book market represents only about 1% of the total LLM text generation opportunity. Even then the total addressable market of LLM text generation is approximately $300 million annually—a modest figure when compared to AWS, which raked in $90 billion in 2023 as the cloud market leader.
While this scenario may seem hypothetical, I’ve witnessed firsthand its implications in various startups. The true value proposition of LLMs lies in their ability to convert unstructured data from sources like websites and documents into structured information with reasonably high accuracy. Yet, the real profit lies in the products built on top of LLM technology.
Consider a healthcare startup leveraging HIPAA-compliant LLMs to automate regulatory form-filling. They charge around $50 per form while paying just about $1 to the LLM provider. Interestingly, their cloud expenditures far exceed their LLM costs, a common trend among many startups. Adding to the challenge is the fact that most major LLM providers (except for Claude) adhere to OpenAI-compatible API standards for text generation. This interoperability allows businesses to switch providers easily by simply adjusting their API endpoints.
So, how can LLM providers monetize effectively? The answer might lie in multimedia applications.
For instance, generating a single 1024x1792 image on OpenAI costs merely $0.12. Extrapolate this to a 1-minute movie at 60 frames per second—an unoptimized 36,000 frames or an optimized 1,000 frames—would cost around $120 to produce. A full-length feature film would bring in approximately $14,400 in revenue. This revenue potential far surpasses the minimal cost of generating a book. Even there, the value capture might end up in the hands of movie studios unless there is a profit-sharing partnership that happens between the LLM provider and the movie studio.