Artificial Intelligence strategic overview

The AI revolution is redefining business, technology, and investment landscape at an unprecedented pace. In our Artificial Intelligence’s latest Models & Architecture Overview, we unveil the structural layers of the AI ecosystem and highlight where innovation and value creation are accelerating the fastest.

This report is grounded in over 50 industry veterans interviews across leading AI companies and infrastructure providers, offering first-hand insight into how AI is transforming industries globally.

‍

The rise of generative and agentic AI

AI has evolved significantly from the early days of rule-based systems to today's Generative AI (GenAI) models that create new content (text, images, video) and agentic AI capable of autonomously performing tasks like emailing clients or managing support tickets. This evolution marks a shift from simple automation to cognitive augmentation.

‍

The next frontier? AI agents that act as digital coworkers. These systems, built on large language models (LLMs), are forecast to grow at 45% CAGR, reaching a $52B market by 2030. Their ability to interpret context, plan actions, and independently execute them is transforming how companies operate from sales to cybersecurity.

‍

Mapping the AI value chain: 5 interconnected layers

The AI ecosystem is structured into five fundamental layers:

• Applications – End-user tools like AI-based supply chain planning tools, ChatGPT or AI-driven marketing assistants.

• Platforms – Enable AI development without starting from scratch (e.g., Amazon Bedrock).

• Models – The foundational AI brains (e.g., GPT-4, Mistral, Claude).

• Infrastructure – Cloud and compute environments where models are trained and served.

• Hardware – Specialized chips (notably GPUs) powering everything beneath.

‍

Each layer plays a crucial role, but increasingly, players are moving across boundaries, model builders launching apps, infrastructure firms offering platforms, and everyone racing to secure hardware supply.

A fragmented yet strategic landscape

Despite the buzz, the AI landscape remains highly fragmented, particularly in the application and platform layers. By contrast, the model layer is dominated by a handful of players, including hyperscalers like OpenAI, Google, and Anthropic.

This fragmentation creates both risk and opportunity. Integration and interoperability challenges persist, but so do avenues for product differentiation and value creation. ‍

‍

Investment perspectives across layers

Not all parts of the value chain carry the same profile for investors:

• High-risk/high-reward: Applications, platforms, and models where innovation moves fast but competition is fierce.

• Defensive positions: Hardware where deep R&D moats and capital intensity offer more protection but less upside.

‍

The report emphasizes that infrastructure and model builders are increasingly vertically integrating, securing their own GPU supply and moving upstream to capture more of the value chain.

‍

Where AI creates value: 3 dimensions of impact

AI is not just a tool; it’s a business accelerator. We identify three key vectors of AI-driven value creation:

• New revenue generation (e.g., ability to enlarge addressable segments/markets such as long-tail of SMBs).

• Product differentiation (e.g., AI-enhanced compliance tools).

• Operational efficiency (e.g., AI reducing marketing campaign timelines by over 90%).

‍

In every case, AI empowers companies to do more, faster and smarter.

‍

To understand how AI is reshaping industries, and where the most promising opportunities lie across the value chain, request full access to the report.