Unpacking the AI Economy: Part 2

How Specialist Partnerships Will Dominate Enterprise AI

This is Part 2 of a two-part post. You can find Part 1 here.

As the AI market continues to evolve, a clear shift is emerging in how success will be defined and where it will come from. While mega-companies like NVIDIA, OpenAI, and Google have been essential in seeding the market, to this point dominating every layer of the value chain, the next wave of growth can only come through specialisation.

---

---

The future economic model of the AI ecosystem will likely favour smaller, specialised players that can excel in specific niches. This trend was evident in many of the announcements at CES in Las Vegas last week. As this shift takes hold, it will pave the way for a new generation of innovators to step forward.

This shift reflects a broader trend in technology markets: specialisation accelerates innovation. Consider how Tesla disrupted traditional automakers and drove the adoption of electric vehicles. In the AI economy, success will go to companies that understand and address the market's nuanced demands.

Recognising specialisation, effectively mapping the ecosystem, and developing precise strategies will be crucial for staying ahead in the market. Anything less puts companies at risk of being overtaken by a diverse and relentless wave of competitors.

The future of the AI economy will increasingly depend on partnerships between large players and smaller, specialised companies. By collaborating with these smaller innovators, the big players can tap into niche expertise and specialised capabilities that they may not possess internally. This shift in focus, away from traditional big company-to-big company partnerships, will foster a more resilient, diverse, and progressive ecosystem. The real growth will come from creating a network of interdependencies where larger companies enable and empower smaller players to drive innovation into the end-user customer base, making the entire ecosystem stronger and more dynamic.

This kind of decentralised structure is not new. We've seen it with the internet, open source, and cloud computing. True success doesn’t come from consolidating control, but from fostering segmentation and specialisation, where a diverse group of players collectively drive growth. A dominant AI monopoly is limited. It is just a piece of the pie. But a dynamic, specialised economic ecosystem is where the real opportunity lies. It expands the pie…exponentially.

To succeed and dominate, technology vendors must understand the critical interdependencies within the ecosystem. Vendors like Microsoft, ServiceNow, Amazon, and Salesforce must position themselves as indispensable hubs that connect and enhance the various components of the ecosystem. For example, these Enablers (see Part 1), will play a central role in driving the growth and functionality of agentic AI, not through alliances, but only by enabling other smaller specialists to innovate and build on their platforms.

I believe this will be the area of greatest growth in 2025 and ultimately set up 2026 as a watershed year for AI transformation in the enterprise.

AI Vendors You Are Doing Great, But Keep Going

As I touched on in part one, while many companies are doing impressive things with AI, most have yet to clearly define and fully express their role within the broader ecosystem. As a result, what we've witnessed so far is more of a chaotic scramble for market share rather than a coordinated effort to contribute meaningfully to the ecosystem.

Offerings like Microsoft CoPilot, ServiceNow's Now Assist, and even their collaborative AI Assistant partnership are undeniably powerful and innovative. Yet, they remain limited by a fundamental issue. That is, they primarily combine components from the same part of the ecosystem rather than creating entirely new connections across segments.

These products demonstrate exceptional execution within their respective domains, but they largely operate within the confines of Enablers of enterprise software platforms. They refine and enhance workflows, improve user experiences, and boost productivity but I don’t think they are yet reshaping the fundamental possibilities of AI.

To truly unlock the transformational potential of AI, it will take much more than incremental improvements or aligning offerings from similar players within the same component part of the market. It will require cross-ecosystem innovation: the ability to merge capabilities from entirely different segments of the AI economy.

What companies like Microsoft, ServiceNow, and Salesforce have achieved with AI is undeniably impressive. However, the focus needs to shift, or they risk spending vast amounts on broad marketing efforts that only optimise legacy systems, rather than unlocking AI's true transformative potential.

Without a shift in focus, even the most impressive AI products risk being perceived as incremental upgrades rather than the foundation of a truly revolutionary ecosystem. Marketing efforts should align with this ambition. It should not be about efficiency dividends and showcasing what AI can do today. It should be positioning it as the catalyst for tomorrow's transformative breakthroughs.

To achieve this, AI Enablers must form specialised partnerships to tackle high-value challenges across the industries and sectors in which they wish to actively engage; be it healthcare, finance, logistics, and government.

For example, partnerships with suppliers focused on developing game-changing industry-specific large language models (LLMs) are key. Currently, many AI initiatives are focused on improving existing systems. They are prioritising efficiency over transformation.

Partnerships that integrate foundational technologies like LLMs with industry-specific applications will not only refine outdated systems, but will fundamentally disrupt the 20th-century business models on which those industries have been built.

So Where Are We At Today

Right now, many AI products are simply leveraging the strengths of their existing ecosystems without looking outward. CoPilot integrates seamlessly into Microsoft’s software suite, and Now Assist optimises ServiceNow’s workflows. These are excellent demonstrations of AI’s capability, but they are not yet transformative because they rely on refining the status quo rather than challenging it.

To truly revolutionise industries, the enablers must start bridging gaps across the ecosystem. The players that can look beyond their own boundaries and integrate technologies, data, and expertise from other parts of the ecosystem will be the ones to define the next wave of innovation.

By treating the entire market as a singular, unified entity, vendors looking to dominate the AI market are at risk of making a strategic miscalculation. This oversimplification distorts the reality that the AI economy is an ecosystem of interdependent, highly specialised players, each with its own role and competitive strengths.

We’ve seen this before. In the early 2000s, IBM underestimated the rise of Indian outsourcing firms, viewing them as a monolithic threat rather than distinct brands with individual strategies. Companies like Infosys, Wipro, and TCS weren’t simply part of a single "outsourcing wave"; they were specialists, each addressing specific customer needs and industry demands.

By the time IBM recognised the need to deal with them as individual competitors in the new global services economy it had already lost significant ground to them in aviation, banking, and other industries¹. Others will make this mistake with “AI”.

While we may not yet have seen the companies that will ultimately shape AI's future, one thing is clear: specialised startups, niche tech firms, and non-traditional service providers hold the key to unlocking AI's true potential.

In fact, today’s leading AI vendors are depending on it.

---

1. I wrote about this more in The Great Gen AI Race.