Striking It Rich in the AI Gold Rush

By: Damien Kopp

In early 1853, a young German immigrant made his way to California with a head full of dreams.

No, he wasn’t thinking about the gold he would dig up from the Sierra Nevada range.

His goal was more straightforward: make money off thousands of diggers by selling them durable trousers.

This man was Levi Strauss, an opportunist who saw real value in supplying the “picks and shovels” during the California Gold Rush. While most gold diggers found little more than dust, Strauss managed to build a denim empire that endures to this day.

Our modern rush toward artificial intelligence features similar risk and reward profiles. While AI end users across industries hope to make it big, other businesses consistently strike it rich by selling the infrastructure (cloud, data, semiconductors, etc.) to these users.

The AI Value Chain

My analysis of the risk, reward, and market potential of key industry players helped me devise the AI Value Chain. This framework provides insight into who benefits the most from AI development, deployment, and adoption across industries.

In this article, I’ve grouped the value chain into four basic archetypes. Each plays a distinct role in today’s ecosystem: some build the foundations, some innovate at the frontier, and others try to extract value through real-world applications.

Each faces a different balance of risk and reward.

Infrastructure Builders (low risk, high reward)

These are the companies supplying the core building blocks of the AI economy. From cloud and data providers to chip manufacturers and energy suppliers, these players power nearly every stage of AI development and use.

Think NVIDIA, AWS, and Google Cloud. The boom in generative AI has been especially beneficial for these companies, with each of them witnessing a massive surge in data centre revenue.

Similarly, the massive power consumption of large AI models has created significant opportunities for energy suppliers. From Microsoft investing $1.6 billion to re-open a nuclear plant to Google penning an agreement with a nuclear energy company, renewable energy is in particularly high demand as businesses race to find cleaner, greener solutions to power their models.

Infrastructure builders stand to reap the most rewards while facing minimal risk. After all, their role is foundational. Their revenue will keep up with the ever-growing demand for AI capabilities unless viable substitutes emerge. Think about how DeepSeek sunk NVIDIA’s stock by demonstrating it could achieve impressive GenAI results without relying on the chipmaker!

Model Builders/Researchers (high risk, high reward)

This group includes the teams developing the foundation models that underpin today’s AI landscape. Some of the most well-known players include OpenAI, Anthropic, and DeepMind.

Their work is resource-intensive and deeply experimental. As a result, they bear high research and reputational risk, navigate intense regulatory scrutiny, and face significant infrastructure costs.

But when these companies’ models become standards, they reap massive influence and licensing opportunities. The market for large language models (LLMs) especially, like ChatGPT and Claude, is advancing at breakneck speed due to rising consumer interest in GenAI.

App Builders (medium risk, high reward)

App developers create AI-powered software solutions tailored to specific business needs, spanning industries from healthcare to retail. This sector is characterized by high levels of innovation, with companies like Salesforce, Adobe, and UiPath leading the charge.

However, these businesses face medium risk levels due to challenges in achieving product-market fit and differentiating their offerings in a competitive market.

Despite these challenges, the rewards can be substantial. Notion is a great example of successfully riding the AI wave: the productivity startup hit a $10 billion valuation by moving early on AI tools.

End Users (high risk, high reward)

These are the organizations integrating AI into their core operations, including banks, hospitals, logistics firms, and creative agencies.

They face the challenge of real-world implementation: adapting legacy systems, retraining staff, ensuring compliance, and managing risk to customers.

But they also stand to gain transformative rewards like operational efficiency, innovation, and competitive edge if they can truly make AI work at scale.

For instance, consider the contrasting cases of McDonald’s and Wingstop. While the former had to shut down its AI-powered drive-thru ordering system after numerous complaints, the latter’s voice AI pilot was successful enough to start handling 10% of its orders.

Dig for Gold…or Sell Shovels?

The AI boom is reshaping how industries build, compete, and deliver value. Whether you’re developing foundation models, integrating AI into your products, or exploring its potential for your organization, the question remains:

Are you digging for gold… or selling the shovels?

Not everyone can be an infrastructure giant or a model builder. But end users, despite facing the highest risk, can still gain a lot if they play their hand strategically.

Here are a few ways end users can maximize returns while managing risk:

• Start with strategic use cases. Focus on high-impact, low-regret applications that can make the most of agentic solutions. For example, investing in conversational AI is a proven way to reduce call handle times and improve human agent efficiency.
• Prioritize integration over novelty. Don’t chase hype. Focus on integrating AI meaningfully into existing workflows rather than adopting flashy tools that don’t solve real problems.
• Invest in upskilling. Equip your teams with AI literacy. The better your people understand AI’s potential and limits, the smarter your adoption will be.
• Build guardrails early. Establish clear governance, auditability, and ethical standards from the start. This is the foundational layer of the AI Adoption Pyramid, which outlines a three-tiered approach to successful AI integration. Setting acceptable levels of risk early helps you innovate and adapt even when things go south.
• Partner wisely. Choose vendors and platforms that align with your compliance, data security, and transparency needs. A big-name vendor may have the brand appeal, but might not be the best fit for your unique goals and requirements. Involve key stakeholders and spend time analyzing your risk profile to select the right partner.

Final Thoughts

The AI value chain is a diverse and dynamic ecosystem, with each player contributing uniquely to its growth.

From semiconductor manufacturers to end-users, every segment plays a critical role in shaping the future of AI.

By understanding the risks, rewards, and market potential of these industries, businesses and investors can navigate the complexities of the AI landscape and unlock its transformative potential.

So, ask yourself: what are you really building? A one-time bet on AI, or something that will endure beyond the rush?

Who is Damien?

Damien is a Fractional technology leader and innovator at MISSION+. With over 22 years of experience in digital innovation, product strategy, and technology consulting across Europe, North America, and Asia, Damien understands what it takes to drive meaningful change by making the most of technology and talent. Feel free to discuss your AI adoption plans with Damien by reaching out at hello@mission.plus.

Originally published at https://www.koncentrik.co/p/the-ai-risk-reward-model on 26 November 2024.