On Tuesday, I attended an interesting conference where, for the first time, finance experts openly discussed what sounded like an AI bubble. The conversation centered around the meteoric rise of generative AI and posed a critical question: will these companies still be around in five years?
It was unclear if these views were intended to provoke or reflect genuine concern, but I agree that AI's potential blinds many investors. At the same time, I believe the real winners in generative AI are the platforms with 20+ years of user data—and here's why.
Many of you may have missed it, so here it is: X has modified its privacy policy to permit third-party access to user data for AI model training. This change took effect on November 15, 2024. The policy now explicitly states that X can share user information with external companies for AI development.
Why is this so important? The major platforms have decades of images and user data that make training, large language models, or LLMs fast and easy. So, in reality, most of these platforms will become parity over time. Their LLMs are all trained on the same data. On the other hand, Google has taken a different stance, prohibiting third parties from using its data for generalized AI models.
This positions major platforms with vast data reserves, Meta, Amazon, LinkedIn, X, Microsoft, and Reddit, to name a few, at a distinct advantage, making it incredibly difficult for new entrants to compete.
For investors rushing to fund generative AI startups, this could mean trouble. Many of these companies cannot compete in the longer term.
This brings me to another point: there are two forms of AI: Vertical AI and Horizontal AI.
Vertical AI focuses on specific industries or use cases designed to improve performance, develop creativity, or do a task within that vertical. Think about it this way: Gemini from Google is designed to improve Google. Grok intends to improve X's experience. The list goes on.
And then there is what I call "Horizontal AI."
Horizontal AI operates across industries, providing tools to solve diverse problems. Think of it as the connective tissue that helps businesses organize data, enhance workflows, and accelerate decision-making.
Chat GPT defines Horizontal AI this way:
Horizontal AI refers to artificial intelligence systems designed to perform a broad range of tasks across multiple industries, domains, or applications. Unlike Vertical AI, which is tailored to address specific problems within a particular industry (e.g., AI for healthcare or finance), Horizontal AI provides generalizable capabilities that can be adapted to various contexts.
Key Characteristics of Horizontal AI:
General Purpose: Horizontal AI systems are not confined to a single use case but offer flexible tools applicable to diverse problems (e.g., natural language processing, computer vision, or predictive analytics).
Scalability Across Domains: These systems can be integrated into workflows in multiple industries without significant customization, making them highly versatile.
Adaptability: Horizontal AI frameworks often leverage transfer learning or other adaptable methodologies, enabling them to learn from one context and apply insights to another.
Core Technologies: Examples include AI platforms offering tools like
Chatbots or virtual assistants for customer support in various industries.
Recommendation engines for e-commerce, streaming, and more.
Data analysis and visualization tools applicable in marketing, finance, and operations.
We are just starting to see examples of horizontal AI, primarily in what you're hearing about agentic AI. These are the companies where early-stage investors should focus today. These are the companies that are defining strong solutions that use AI to accelerate business outcomes.
ChatGPT: Can assist with writing, brainstorming, or answering questions in many fields.
Google Cloud AI: Provides APIs for language understanding, vision, and machine learning that can be implemented across industries.
mktg.ai: Organizing and analyzing creative assets empowers marketers to unlock the full potential of their campaigns, bridging gaps caused by fragmented AI adoption.
Horizontal AI Is the Future
As AI continues to fragment business processes and channels, Horizontal AI will play a pivotal role in integrating and streamlining these systems.
Why Horizontal AI Matters:
Efficient Deployment: Organizations can adopt Horizontal AI without needing specialized AI for each department.
Cost-Effective: A single system serves multiple purposes, reducing the need for niche solutions.
Encourages cross-industry innovation by sharing AI advancements universally.
The generative AI bubble might be forming, but the Horizontal AI leaders of tomorrow will be the ones who deliver sustainable, transformative value—across industries, and over time.
And we could not be happier to have mktg.ai viewed through this horizontal lens. Reach out to learn more.
Comments