Chinese AI startup DeepSeek has claimed a 545% profit margin for its AI models, but this figure is based on theoretical revenue rather than actual earnings. The company stated that if all AI usage over a 24-hour period had been billed under its R1 model pricing, it would have earned $562,027, with GPU costs of only $87,072. However, actual revenue is lower due to nighttime discounts, lower-priced models, and free user access.
DeepSeek gained prominence in January after launching an AI model that rivaled OpenAI’s GPT-4 (o1 variant) and briefly surpassed ChatGPT on Apple’s App Store rankings. Its claim of potential profitability challenges the notion that AI businesses struggle financially, as major tech companies like OpenAI and Google have yet to prove AI chatbots' sustainability at scale.
Despite its impressive cost efficiency, DeepSeek's claim raises questions about AI monetization and whether startups are truly on the verge of massive profits or simply making optimistic projections.
Read More.
Top comments (0)