The Future of AI: Function Calling, API Updates, and Contrarian Theses

The Future of AI: Function Calling, API Updates, and Contrarian Theses

Introduction:
Artificial Intelligence (AI) is rapidly evolving, bringing about significant changes in various industries. In this article, we will explore the latest updates in function calling and API advancements, as well as delve into five contrarian AI theses that challenge conventional wisdom. By understanding these developments, we can gain valuable insights into the future of AI and its impact on businesses and investors.

Function Calling and API Updates:
One of the notable updates in AI is the introduction of the new 16k context version of gpt-3.5-turbo. This version offers a significant improvement over the standard 4k version, allowing for more efficient function calling and enhanced API capabilities. With this update, developers can expect a 75% cost reduction on the state-of-the-art embeddings model, making AI more accessible and affordable for a wide range of applications.

Furthermore, there is a 25% cost reduction on input tokens for gpt-3.5-turbo, enabling developers to process larger volumes of data without incurring exorbitant expenses. The text-embedding-ada-002 model, known for its popularity, now comes at a reduced cost of $0.0001 per 1K tokens. These advancements in function calling and API updates pave the way for more efficient and cost-effective AI integration.

Contrarian AI Theses for Early Stage Investors:

1. Horizontal LLMs will lose: Many AI experts believe that Language Models (LLMs) alone will not lead us to Artificial General Intelligence (AGI). Instead, the next breakthroughs in AI will come from new technologies rather than merely accumulating more data. Additionally, the thesis suggests that using multiple versions of vertical LLMs might be more beneficial in most applications, as opposed to relying on a single horizontal LLM.

2. AI's selective impact on companies: Unlike previous technological waves, AI will not revolutionize entire markets but will impact specific companies. AI applies intelligence and learning to different steps in the corporate value chain, meaning that companies with similar value chain structures are likely to benefit the most. This challenges the notion that AI's impact can be generalized across industries and emphasizes the importance of understanding each company's unique workflow and value chain.

3. The demise of competitive advantage: The rise of AI will diminish traditional forms of competitive advantage. A well-known example is Google's "we have no moat" memo, which highlights the belief that AI will level the playing field by enabling widespread access to advanced technologies. This suggests that companies need to adapt and innovate continuously to stay ahead in an AI-driven economy.

4. Bifurcation of the economy: AI will divide the economy into two worlds - the real world and the AI world. As AI agents become more prevalent, they will replace traditional software application interfaces, leading to a collapse of customer acquisition channels. This shift necessitates a reevaluation of business strategies, particularly for companies relying on product-led growth or community-driven go-to-market approaches.

Actionable Advice:

1. Embrace vertical LLMs: To leverage the full potential of AI, consider using multiple versions of vertical LLMs tailored to specific applications. This approach allows for more targeted and efficient AI integration, leading to better results and competitive advantages.

2. Adapt and innovate: Recognize that AI is reshaping industries and challenging established norms. To thrive in this evolving landscape, companies must embrace change, continually innovate, and optimize their workflows to incorporate intelligent and learning-based systems.

3. Anticipate the AI-driven economy: Prepare for the bifurcation of the economy by exploring the potential of AI agents and their impact on customer acquisition channels. Rethink your company's go-to-market strategies to align with the changing dynamics of the AI world.

Conclusion:
As AI continues to advance, function calling and API updates play a crucial role in making AI more accessible and cost-effective. Additionally, contrarian AI theses challenge traditional notions and offer valuable insights for early-stage investors. By embracing vertical LLMs, adapting to AI-driven changes, and anticipating the future economy, businesses can position themselves for success in an increasingly AI-driven world.