The intersection of AI and business growth is an area of great interest and potential. As AI continues to advance and become more accessible, it is clear that it will have a profound impact on various industries and business models. In this article, we will explore six new theories about AI and how they relate to the three horizons of growth framework.

The intersection of AI and business growth is an area of great interest and potential. As AI continues to advance and become more accessible, it is clear that it will have a profound impact on various industries and business models. In this article, we will explore six new theories about AI and how they relate to the three horizons of growth framework.

Theory 1: Fine-tuned models win battles, foundational models win wars

When it comes to AI, there are two types of models: foundational models and fine-tuned models. Foundational models are broad in scope and aim to perform a specific task well, such as text generation. On the other hand, fine-tuned models are tailored to a specific use case, like healthcare text generation. While fine-tuning may not necessarily outperform foundational models, it can significantly lower the cost of prompt completion for narrow use cases. Over time, fine-tuned models can gradually improve, while foundational models experience step-changes in performance.

This theory aligns with the three horizons framework by highlighting the importance of both core businesses (horizon one) and emerging opportunities (horizon two). Foundational models can be seen as the core businesses, providing the greatest profits and cash flow. On the other hand, fine-tuned models represent the emerging opportunities that have the potential to generate substantial profits in the future.

Theory 2: Long-term model differentiation comes from data-generating use cases

Data loops, which involve building feedback mechanisms into AI products and using that feedback to retrain the model, are crucial for long-term model differentiation. Startups that can capture the model-to-output loop and continuously retrain their models have the potential to create specialized winners. This theory suggests that owning both the model provider and the endpoint solution is necessary for successful long-term differentiation.

In the context of the three horizons framework, this theory aligns with horizon three, which encompasses ideas for profitable growth down the road. Startups that focus on data-generating use cases and invest in building feedback mechanisms into their products are likely to thrive in the long term.

Theory 3: Open source makes AI startups into consulting shops, not SaaS companies

Open source AI models have the potential to disrupt the AI market by putting downward pricing pressure on model providers that sell access to their models via API. When competing with free, AI startups are often forced to compromise by being cheap. However, this theory suggests that open source AI models also create opportunities for AI startups to provide consulting services instead of traditional software-as-a-service (SaaS) offerings.

This theory challenges the traditional notion of SaaS companies and aligns with the three horizons framework by highlighting the need for startups to adapt their business models (horizon two) and explore new avenues for growth (horizon three). AI startups that can provide consulting services based on open source AI models have the potential to tap into emerging opportunities and achieve profitable growth.

Actionable Advice:

1. Embrace fine-tuned models: Consider the potential of fine-tuning foundational models to lower the cost of prompt completion for narrow use cases. Gradually improving fine-tuned models can provide a competitive advantage in the long term.

2. Invest in data-generating use cases: Focus on building feedback mechanisms into AI products and continuously retraining the models. This will facilitate long-term model differentiation and set the stage for profitable growth.

3. Explore consulting opportunities: Instead of solely relying on traditional SaaS models, consider providing consulting services based on open source AI models. This can open up new avenues for growth and help navigate the challenges posed by open source AI.

In conclusion, the rapid advancement of AI presents both challenges and opportunities for businesses. By understanding and leveraging the theories discussed in this article, companies can position themselves for growth across the three horizons. Embracing fine-tuned models, investing in data-generating use cases, and exploring consulting opportunities are actionable steps that can lead to success in the AI-driven future.