AI Stock Performance - H1 2026

I previously wrote about Jensen Huang’s description of AI as a five-layer cake, where each layer depends on the one below it:

𝐄𝐧𝐞𝐫𝐠𝐲 powers the stack: power generation, transmission, grid storage, on-site power.
↓
𝐂𝐡𝐢𝐩𝐬 convert energy into compute capacity: foundries, GPUs/ASICs, memory, interconnects.
↓
𝐈𝐧𝐟𝐫𝐚𝐬𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞 turns chips into usable scale: hyperscale cloud, GPU clouds, data centers, orchestration, cooling.
↓
𝐌𝐨𝐝𝐞𝐥𝐬 convert compute and data into intelligence: foundation, reasoning, multimodal, small & edge, domain-specific.
↓
𝐀𝐩𝐩𝐥𝐢𝐜𝐚𝐭𝐢𝐨𝐧𝐬 put that intelligence in front of users: consumer, enterprise, coding, creative, healthcare, robotics, industrial.

Looking at H1 2026 stock performance, many of the strongest performers were not application companies. Memory, storage, semiconductor equipment, AI infrastructure, and power-related businesses dominated much of the leaderboard, with some stocks rising more than 7x in just six months.

This reflects where today’s bottlenecks are. A few years ago, the constraints were GPUs and foundation models, which drove significant gains in NVIDIA’s stock and the valuations of private companies like OpenAI and Anthropic. Today, memory, storage, power, and data center capacity have become equally important.

In upcoming posts, I’ll examine each layer in more detail and explore how different parts of the AI ecosystem have performed and why.