Serenity: AI Bottlenecks Could Create Significant Outperformance Opportunities in 2027

The race to expand artificial intelligence infrastructure is becoming more than a contest for computing power. It is also a contest for access to the less visible components that make large-scale systems work.

In a November 15 post, investor Serenity argued that supply constraints in parts of the AI hardware chain could create substantial opportunities for investment outperformance in 2027. The thesis is not simply that demand for AI will rise. It is that a relatively inexpensive component can acquire strategic importance when a shortage threatens to delay a much larger project.

For cryptocurrency investors, the connection is worth examining. AI infrastructure and blockchain are distinct markets, but they increasingly intersect around decentralized computing, tokenized infrastructure projects and competition for data center resources. Understanding the physical supply chain can help investors assess those narratives more carefully.

Why a Small Component Can Have an Outsized Impact

Serenity’s argument draws on a supply-chain lesson associated with Nvidia CEO Jensen Huang’s management approach: securing scarce inputs early can be valuable even when the inputs represent only a small part of a system’s total cost.

Consider a company building a large computing cluster. If one constrained component holds up deployment, the financial impact may extend far beyond the price of that part. Delayed installation can postpone revenue, disrupt customer commitments and leave other expensive equipment unused.

This creates an incentive for buyers to reserve production capacity before a shortage becomes acute. When several major customers pursue the same strategy, demand can move upstream toward specialist suppliers well before it appears in finished-product sales.

Serenity cited SOI as an example, noting an estimate of approximately $400 million in photonics revenue for its 2028 fiscal year. That figure is a forecast, not realized revenue, and should not be treated as a guarantee of future demand or investment performance.

The broader point is that the incentive to secure scarce capacity does not depend on one executive’s continued leadership. It follows from the economics of avoiding potentially costly delays.

Where AI Infrastructure Bottlenecks May Form

Serenity identified several areas that investors are already watching, including those highlighted by Morgan Stanley:

• Copper-clad laminates (CCL): Materials used in the production of circuit boards.
• ABF substrates: Specialized packaging substrates used in advanced semiconductor products.
• Printed circuit boards (PCB): Boards that provide electrical connections between components.
• Optical module components: Parts used in systems that transmit data using light.

These categories are not interchangeable. Each has its own customers, production constraints, competitive landscape and valuation risks. A growing end market does not automatically translate into durable pricing power for every supplier.

Serenity expressed greater interest in niche components that account for a small share of system cost but face meaningful supply limitations. Another area of interest is capacity that a buyer can secure relatively cheaply before demand intensifies.

This is an important distinction for investors: the largest or most recognizable market is not necessarily where scarcity has the greatest effect on returns.

Why Optical Architectures Matter

As AI clusters grow, moving data between chips and systems becomes increasingly important. Optical technologies are one approach to addressing connectivity requirements, although adoption depends on technical performance, cost and the readiness of the wider manufacturing ecosystem.

Two architectures frequently discussed in this context are near-packaged optics (NPO) and co-packaged optics (CPO). In general terms, they place optical components closer to the electronics they serve than in some conventional arrangements. The goal is to improve how systems handle high-speed data movement, though designs and implementation choices vary.

Serenity expects the gradual rollout of these architectures to create opportunities in constrained parts of the supply chain. That outcome remains uncertain: a promising architecture may take longer to commercialize than expected, and demand for a component does not necessarily benefit every company exposed to it.

The wider importance of advanced computing infrastructure is reflected in the International Energy Agency’s analysis of energy demand from data centers and AI. Power availability is only one constraint, but it illustrates why physical infrastructure matters alongside software adoption.

What This Means for Crypto Investors

The connection to blockchain is indirect but relevant. Some crypto projects seek to coordinate computing resources through decentralized networks, while others use tokens to support markets for services such as compute or data storage. Their prospects depend on more than token demand: hardware access, operating costs and reliable service delivery can all matter.

Three questions are especially useful when evaluating the AI–crypto intersection:

1. Does the Project Have Access to the Resources It Needs?

A token associated with AI does not itself provide computing capacity. Investors can examine whether a project discloses how its network obtains hardware, who supplies it and whether users can actually access the advertised service.

2. Who Benefits From Higher Demand?

Growth in demand for compute may benefit hardware manufacturers, data center operators, service providers or network participants in different proportions. A blockchain project’s users may increase without its token necessarily capturing an equivalent share of that value.

Reviewing the token’s intended function, supply schedule and links to network activity is therefore as important as assessing the underlying technology.

3. Can the Network Operate Through Supply Constraints?

If equipment, connectivity or electricity becomes more expensive, decentralized infrastructure providers may face pressure on margins. Projects with transparent operating assumptions and measurable usage may be easier to evaluate than those relying primarily on projected future demand.

These questions are relevant to any AI-linked crypto investment, but they do not imply that Serenity’s view on hardware suppliers is a recommendation to buy related tokens.

The Market May Not Price Fundamentals Immediately

Serenity also noted that deleveraging and pessimistic sentiment can weigh broadly on companies across a supply chain, making it difficult for individual fundamentals to stand out in the short term.

That is a useful caution for crypto markets as well. A sound long-term thesis can coexist with sharp price declines, limited liquidity or valuations that already assume aggressive growth. Conversely, a falling price does not establish that an asset is undervalued.

Serenity’s reflection on 2026 investment performance underscores the need for greater selectivity when choosing 2027 positions. For investors assessing the thesis, key variables include actual capacity additions, customer commitments, commercialization timelines and the price paid for exposure.

The Takeaway

Serenity’s central thesis is that AI supply-chain bottlenecks could create significant opportunities in 2027, particularly where inexpensive but scarce components are essential to much larger systems. NPO and CPO adoption may strengthen that thesis, but the timing and beneficiaries remain uncertain.

For crypto investors, the practical lesson is to look beyond the AI label. The strongest analysis connects a project’s token economics and real usage to the physical resources required to deliver its service.

This article is for informational purposes only and does not constitute investment advice. Forecasts and investment outcomes are uncertain.