AI Bubble
The AI bubble narrative began when Nvidia’s May 2023 outlook revealed an abrupt step-up in demand for accelerated computing, triggering a global revaluation of chips, servers, data centers and AI software. Capital has since crowded into a relatively narrow group of beneficiaries whose valuations discount years of rapid infrastructure spending and eventual productivity gains. The boom can persist while hyperscalers keep raising capital expenditure, but overbuilding, model commoditization, power constraints or weak customer monetization could expose excess capacity and compressed returns. At the same time, generative AI is already pressuring businesses built around legacy processors, outsourced knowledge work, stock content and paid access to information.
How this story could end...
The endings this story genuinely supports, written from the same evidence as the story updates. The value ranges are illustrative examples of how the affected companies as a group could re-rate — not forecasts, targets or advice.
Mapping out the endings…
How the two sides have traded
The bull and bear baskets for this story, equal-weighted and indexed so the two lines start together.
Each scenario's basket — 3 months
Both sides on one chart — 10 stocks on the bull side, 10 stocks on the bear side. Neither side is emphasised.
Each line tracks the 10 companies most exposed to this story, so a broad basket of loosely related names does not flatten it. The full list below is unchanged.
Jul 20, 2026 – Oct 19, 2026 · 69 sessions
Equal-weighted performance of the 10 most exposed stocks over the last 3 months.
Most affected stocks
Bull case
Investor enthusiasm for its AI Platform and rapid commercial growth has made Palantir a prominent software expression of the theme.
Asia Vital Components supplies fans, heat sinks and liquid-cooling equipment needed to manage the rising thermal density of AI servers.
Supermicro assembles dense, liquid-cooled GPU servers, making it a direct beneficiary of rapid AI-cluster deployment but also highly sensitive to an infrastructure slowdown.
Azure AI services, its OpenAI partnership and Copilot products position Microsoft to monetize AI across cloud and enterprise software.
Oracle is investing aggressively in accelerator clusters and can gain cloud share if demand for AI training and inference remains elevated.
Netweb builds high-performance computing systems and AI servers, positioning it for India's sovereign and enterprise compute investments.
The complexity and value of AI accelerators and high-bandwidth memory increase semiconductor testing intensity.
Wistron manufactures accelerator-based servers and benefits from hyperscaler and enterprise orders for AI computing systems.
Dell is a major vendor of AI-optimized servers and integrated infrastructure, making it a direct beneficiary of enterprise and cloud spending on accelerated computing.
Marvell supplies data-center networking silicon, optical connectivity products and custom AI accelerators for hyperscale customers.
Bear case
Generative image models can substitute for licensed stock content and weaken traditional pricing power.
Low-cost generative image and video tools threaten demand and pricing for Shutterstock's conventional stock-content library.
Free and inexpensive AI tutors directly undermine demand for Chegg's subscription-based homework assistance.
The shift from general-purpose CPUs toward accelerated computing has exposed Intel's product and manufacturing disadvantages.
AI tools can automate portions of the writing, design, coding and administrative work that drives activity on Upwork's freelance marketplace.
Generative tutoring and content-creation tools could erode the value of traditional educational courseware and alter demand for Pearson's learning products.
AI-assisted software development and support could compress Wipro's billable labor volumes and accelerate pricing pressure across outsourced IT services.
Low-cost generative image and video tools threaten to commoditize parts of Adobe's creative software proposition despite its own AI products.
Generative AI can automate portions of Genpact's labor-intensive finance, customer-service and back-office outsourcing work, pressuring pricing and headcount growth.
Generative AI threatens to automate portions of TCS's labor-intensive application development, testing and business-process work, potentially pressuring billable headcount and pricing.
Reporting behind this story
Looking at what moved this story…