The Wild West of the Value Chain Beyond Big Tech: 10 Case Studies from AI Infrastructure to Gene Editing

The key to understanding the current market landscape is understanding the underlying business models that operate entirely outside the gravitational pull of Big Tech aggregators. When we examine a seemingly disparate basket of unclassified US equities—a catch-all bucket ranging from artificial intelligence infrastructure to quantum security and biotechnology—we are not just looking at random tickers. We are observing the specialized value chains where the next phase of value capture is taking place. A platform empowers third parties; an aggregator intermediates them. But out here on the fringes, companies do not aggregate massive consumer attention; instead, they seek to monopolize extremely narrow, highly specific chokepoints in the supply chain.

One Stop Systems (OSS.US)

We typically think of major silicon designers when we discuss the AI boom, but the true depth of this structural shift is found much further down the stack. One Stop Systems is operating on the rugged, physical edge of this compute paradigm. The stock has consolidated steadily within its recent range, supported by its Q2 2026 revenue from continuing operations of USD 9.3 million, up 62.3% year-over-year. In the context of Aggregation Theory, OSS is not an aggregator. Instead, it is providing a critical complement—ruggedized edge computing for defense and commercial uses. By securing a USD 8.4 million defense contract and being added to the Russell 2000 index in June 2026, the company is successfully commoditizing its complement and moving up the value chain to capture more margin.

Innodata (INOD.US)

At the data processing layer, Innodata illustrates another fundamental truth about AI: as large language models move relentlessly toward commoditization, high-quality, specialized data becomes the primary differentiating asset. The company recently reported a record Q2 2026 revenue of USD 92.1 million, up 58% year-over-year. More tellingly, they secured two new LLM projects from a large tech client representing USD 44 million in annualized run-rate revenue. This means that data engineering is becoming the new premium layer where real economic rents are extracted.

Cohu (COHU.US)

Similarly, in the physical testing layer, Cohu has demonstrated resilient market performance as its Q2 2026 net sales rose 38% to USD 149 million. By receiving multiple forward-looking orders to test next-generation GaN power devices for AI data centers, Cohu is capturing tremendous value before the final consumer-facing products even reach the hyperscalers.

SEALSQ (LAES.US)

This, though, is exactly backwards if we assume the silicon-based paradigm is static. The ongoing shift toward post-quantum architectures and decentralized networks is unbundling traditional enterprise security models. SEALSQ is heavily investing in post-quantum secure semiconductors. With preliminary H1 2026 revenue surging 120% and a new USD 100 million quantum technology investment phase kicking off, they are betting that physical AI will require entirely new roots of trust. When security cannot be guaranteed by legacy cryptography, owning the hardware-level trust protocol becomes an incredibly valuable chokepoint.

SentinelOne (S.US)

This systemic necessity for upgraded security architecture is also visible in SentinelOne. Outperforming several of its software peers recently, the AI-driven cybersecurity firm reported Q1 FY2027 total revenue of USD 277 million, a robust 21% increase year-over-year. As the digital infrastructure becomes more complex, securing the very data flows that platforms rely upon transforms into a highly defensive and lucrative business.

Hyperliquid Strategies (PURR.US)

Meanwhile, in the decentralized finance space, Hyperliquid Strategies offers a completely different mechanism of value capture. Operating as a digital asset treasury focused on the Hyperliquid ecosystem, the stock has traded at a persistent discount to its net asset value despite holding significant unrealized gains in HYPE tokens this year. This is fundamentally an attempt to create a new financial primitive—an alternative to centralized capital aggregation methodologies.

CRISPR Therapeutics (CRSP.US)

When we look at biology and deep space physics, the dynamics change entirely. Here, there are no aggregators, only pure intellectual property bottlenecks characterized by massive fixed costs and near-zero marginal costs of reproduction. CRISPR Therapeutics is the prime example of an IP monopolist. Its flagship gene-editing therapy, CASGEVY, generated USD 76 million in Q2 2026, jumping 151% year-over-year. Following the recent FDA approval for younger children with SCD or TDT, CRISPR is proving that owning the foundational mechanism of editing translates into durable, long-term value.

China Pharma Holdings (CPHI.US)

China Pharma Holdings is pursuing a more traditional healthcare model, having recently closed a USD 5 million registered direct offering to support the Q1 2025 launch of its dry eye treatment device. It is a calculated play to secure a niche market segment within a highly competitive regulatory environment.

Astrotech (ASTC.US)

On an even more speculative frontier, Astrotech is pivoting entirely from mass spectrometry to lunar resource infrastructure and space-based quantum computing manufacturing. It is a high-stakes, long-term gamble to secure a supply chain chokepoint before the commercial market even physically exists.

Luckin Coffee (LKNCY.US)

The fascinating outlier in this group is Luckin Coffee. Despite operating physical coffee shops, it is structurally acting as a digital aggregator. Reporting Q2 2026 total net revenue of RMB 15.86 billion (USD 2.33 billion)—a 28.5% increase—the company now operates over 36,000 stores globally. By intermediating the consumer relationship entirely through its app, Luckin forces the physical storefront to become a commoditized fulfillment center. However, the recent management and hygiene challenges at the store level highlight the inherent friction of this model: digital aggregation scales infinitely, but human labor deployed in physical spaces does not.

The key takeaway from this diverse cohort is that value creation outside the tech giants looks fundamentally different. It is found in the unbundling of complex systems, whether in edge computing, post-quantum cryptography, or gene editing. This means that investors must look for companies creating narrow but deep moats, which means that the next generation of outsized returns will come from specialized suppliers, which is why the most compelling narratives often hide in the unclassified edges of the market.

This article does not constitute investment advice.