Cerebras CEO on the Future of Data Centres, Token Costs & Memory | Should US Companies Sell to China
TL;DR
Cerebras CEO Andrew Feldman argues AI infrastructure is not a bubble because supply is struggling to catch up with explosive demand, evidenced by $25 billion in data center backlogs and severe memory shortages. Cerebras' recent IPO reflects this reality, with the company positioned advantageously using SRAM instead of scarce HBM memory while industry-wide compute costs continue their historical decline.
🏗️ The 'No Bubble' Infrastructure Reality 3 insights
Infrastructure chases demand rather than leading it
Unlike fiber optic or railroad bubbles where infrastructure preceded demand, AI data centers face a $25 billion backlog because supply cannot keep up with current needs.
Data center metering prevents market oversupply
Construction delays and permitting challenges act as necessary traffic meters that smooth demand and prevent the market from gorging on excess capacity too quickly.
2025 marked the usefulness inflection point
AI models crossed a threshold in early 2025 to become genuinely useful tools rather than novelties, driving exponential demand across all demographics from Silicon Valley to 85-year-olds.
⚡ Supply Chain Constraints & Cerebras' Advantage 3 insights
HBM memory shortage will persist for years
With only three suppliers (Samsung, Micron, Hynix) enjoying 80-85% gross margins, high-bandwidth memory shortages will continue as building new fab capacity requires $40 billion and five years.
Cerebras avoids critical supply bottlenecks
Unlike GPUs, Cerebras chips use SRAM etched by TSMC during logic manufacturing, avoiding both the HBM shortage and CoWoS packaging constraints while utilizing available 5nm capacity.
GPU prices skyrocketing amid component scarcity
Supply constraints have driven GPU prices through the roof, while Cerebras maintains cost advantages by not paying premiums for HBM or advanced packaging.
🎯 Market Strategy & Competitive Dynamics 3 insights
Nvidia's Neo cloud strategy creates dependence
Nvidia has funded and over-allocated to Neo clouds to create hyperscaler competitors, fostering an unhealthy dependence while hyperscalers retain advantages in security and software ecosystems.
Vertical integration limits hardware market size
Google's full-stack ownership from TPU to data center constrains their hardware market to internal demand only, historically limiting volume and cost reduction compared to merchant silicon vendors.
OpenAI forced into disadvantageous supply deals
Facing supply constraints, OpenAI purchased down-revision H100s from Elon Musk rather than current-generation B200s, leaving them one to two generations behind despite early contracting efforts.
📉 The Trajectory of Compute Costs 2 insights
Cost per compute will continue historic decline
Despite current shortages, the industry will see massive reductions in cost per unit compute over three to four years as all chipmakers improve designs to deliver more tokens per dollar and watt.
Performance gap expected to widen significantly
Cerebras currently delivers 15x faster performance through architecture alone and expects this advantage over GPUs to increase as designs evolve.
Bottom Line
The AI infrastructure buildout is supply-constrained rather than a demand bubble, meaning companies that secure compute capacity now—even if not the latest generation—gain competitive advantage, while specialized architectures that bypass HBM and CoWoS bottlenecks are positioned to capture disproportionate value during the shortage.
More from 20VC with Harry Stebbings
View all
Why Anthropic Are Causing a Comp Crisis & Why You’d Never Hire From Salesforce or ServiceNow
Former Snowflake CRO Chris Degnan and sales leader Chad Pet explain why Anthropic's massive compensation packages are distorting the market, detail why Salesforce and ServiceNow veterans make poor hires (having never opened new logos), and emphasize that only booked annual contracts—not usage metrics—create durable revenue.
Andrej Karpathy Joins Anthropic | SpaceX Files S1: How Does it Trade | Cerebras Smashes Day 1
The episode breaks down Anthropic's staggering $900 billion valuation and Andrej Karpathy's addition to the team, contrasting its clean fundraising style with OpenAI's complexity. The hosts debate whether enterprise AI spending (exemplified by Salesforce's $300M Anthropic contract) can 4x to justify these valuations, or if improving efficiency and cheaper agents
The One Man Accelerator at The Four Seasons & Why VCs Can Be Sharks | Josh Browder
Josh Browder explains his unique 'one-man accelerator' model where he houses young founders in Four Seasons residences while investing at sub-$5M valuations to help them avoid the three fatal pre-seed traps: running out of money, hope, or co-founder trust. He shares specific heuristics for identifying authentic founders with deep problem connections versus 'ideological frauds' who reverse-engineer founder archetypes using AI.
The Five Year Desert to Product Market Fit and a $5.3BN Valuation | Shiv Rao, Founder @ Abridge
Abridge founder Shiv Rao details the company's five-year desert period before achieving product-market fit and a $5.3B valuation, explaining how unwavering conviction in their healthcare conversation thesis combined with flexible execution allowed them to capitalize on the 2023 LLM wave and clinician burnout crisis.