The Future of Generative AI: Ownership, Challenges, and Opportunities

Hatched by Kazuki
Aug 16, 2023
4 min read
7 views
Copy Link
The Future of Generative AI: Ownership, Challenges, and Opportunities
Introduction:
Generative AI has witnessed remarkable growth, with various applications such as image generation, copywriting, and code writing already surpassing $100 million in annual revenue. While infrastructure vendors have emerged as the biggest winners in this market, other players, including application companies and model providers, are also vying for a piece of the pie. In this article, we will explore the ownership dynamics, challenges faced by different players, and the potential opportunities that lie ahead in the generative AI landscape.
Infrastructure Vendors: The Power Players:
Infrastructure vendors have established themselves as key players in the generative AI market. They capture a significant portion of the revenue flowing through the stack. With app companies spending a significant percentage of their revenue on inference and fine-tuning, it is estimated that around 10-20% of the total revenue in generative AI goes to cloud providers. Among these providers, Nvidia stands out as a major winner, reporting billions in data center GPU revenue, a substantial portion of which comes from generative AI use cases.
Challenges Faced by Application Companies:
While application companies experience rapid revenue growth, they often struggle with retention, product differentiation, and gross margins. The absence of strong technical differentiation necessitates the exploration of alternative strategies for long-term customer value. B2B and B2C apps can leverage network effects, data ownership, and complex workflows to drive differentiation and enhance customer retention. It is crucial for these companies to focus on improving margins and differentiating themselves in the market.
Model Providers: Commercialization and Hosting:
Model providers, responsible for the existence of the generative AI market, have not yet achieved large-scale commercial success. However, the demand for proprietary APIs is growing rapidly, and hosting services for open-source models are emerging as convenient hubs for model sharing and integration. Hosting plays a vital role in the commercialization of generative AI models, as it provides a platform for model producers and consumers to connect. Additionally, many model providers have embraced the concept of public good and have incorporated it into their mission, without hindering their fundraising efforts.
The Ownership Dilemma:
The question of whether most model providers actually want to capture value arises. Some model providers have organized as public benefit corporations (B corps) or adopted profit-sharing schemes, prioritizing the public good over maximizing profit for shareholders. This dilemma raises a reasonable discussion about the motivations and goals of model providers. While the promise of generative AI is immense, there is a need to strike a balance between capturing value and ensuring responsible use.
The Road Ahead: Challenges and Opportunities:
The generative AI landscape faces both challenges and opportunities. Margins are expected to improve as competition and efficiency in language models increase. The departure of AI tourists from the market is likely to enhance customer retention. Vertically integrated apps hold an advantage in driving differentiation, while horizontally integrated companies can succeed in offering AI as part of a larger feature set. The future of generative AI may not witness a winner-take-all dynamic but rather a mix of successful vertical and horizontal approaches, depending on the end-markets and end-users.
Actionable Advice:
- 1. Differentiate through Network Effects and Data Ownership: B2B and B2C app companies should focus on building network effects and retaining ownership of valuable data. This can help drive long-term customer value and enhance differentiation in a competitive market.
- 2. Embrace Hosting Services for Model Sharing: Model providers should consider leveraging hosting services for open-source models to facilitate easy sharing and integration. This can create indirect network effects between model producers and consumers, fostering collaboration and commercialization.
- 3. Balance Value Capture with Responsible Use: Model providers need to navigate the ownership dilemma by finding a balance between capturing value and ensuring responsible use of generative AI. Incorporating the public good explicitly into their mission can be a strategic move without hindering fundraising efforts.
Conclusion:
Generative AI has revolutionized various industries, with applications such as image generation, copywriting, and code writing already generating significant revenue. Infrastructure vendors, application companies, and model providers all play crucial roles in this evolving landscape. While infrastructure vendors dominate the revenue flow, application companies and model providers face challenges in retention, differentiation, and commercialization. By leveraging network effects, data ownership, and hosting services, these players can navigate the market successfully. The future of generative AI is likely to witness a mix of vertical and horizontal approaches, creating a dynamic and diverse ecosystem.
Resource:
Copy Link