An Interview with OpenAI President Greg Brockman: ‘There’s not going to be enough compute’

Tae Kim · Key Context by Tae Kim · April 24, 2026 at 02:48 · ⏱ 10 min read  | Read on Substack ↗
Summary
OpenAI's GPT-5.5 launch and Brockman's interview underscore an accelerating demand for AI compute that far outstrips supply, driven by agentic coding and knowledge-work automation. This compute scarcity thesis directly supports semiconductor and memory suppliers, making the article a bullish signal for chipmakers and HBM producers.
  • •GPT-5.5 outperforms Anthropic's Opus 4.7 on reasoning and autonomy tasks, and OpenAI expects significantly faster improvement pace going forward.
  • •Greg Brockman states 'there’s not going to be enough compute in the world to meet the demand' and notes internal compute scarcity has already forced tough allocation decisions at OpenAI.
  • •Agentic coding tools (Codex) are making developers 2–3x faster, and adoption is spreading to non-technical roles like comms and event planning.
  • •OpenAI uses Nvidia as its primary compute partner but also has deals with AMD, Cerebras, and is developing its own custom chip.
  • •Brockman believes 8 billion GPUs (one per person) is not achievable near term; current buildout is on the order of tens of millions of GPUs.
  • •The article's 'Tae’s Take' explicitly calls Nvidia, Samsung/SK Hynix/Micron, Intel, and AMD as beneficiaries of the compute shortage.
Read time 10 min
Length 10,811 chars
Category finance
Trade Ideas
Tae Kim Senior writer, Barron's; author of The Nvidia Way
The article highlights HBM memory suppliers (Samsung, SK Hynix, Micron) as direct beneficiaries of compute scaling. HBM tightness is repeatedly cited in supply-chain discussions, and Micron is the US-
The article highlights HBM memory suppliers (Samsung, SK Hynix, Micron) as direct beneficiaries of compute scaling. HBM tightness is repeatedly cited in supply-chain discussions, and Micron is the US-based proxy for HBM demand. Risk: HBM pricing volatility and potential oversupply if memory makers overinvest; Micron’s HBM3E qualification timelines could disappoint.
MU
Tae Kim Senior writer, Barron's; author of The Nvidia Way
Brockman calls Nvidia 'our closest partner' and Jensen Huang praised GPT-5.5 as a validation of agentic AI. The compute shortage directly drives GPU demand from OpenAI and the broader industry.
Brockman calls Nvidia 'our closest partner' and Jensen Huang praised GPT-5.5 as a validation of agentic AI. The compute shortage directly drives GPU demand from OpenAI and the broader industry. Risk: Potential slowdown in model scaling or shift to custom chips could reduce Nvidia’s share of OpenAI’s compute spend.
NVDA
Tae Kim Senior writer, Barron's; author of The Nvidia Way
OpenAI explicitly uses AMD chips alongside Nvidia to diversify supply, and the overall compute shortage benefits all GPU/CPU suppliers that can fill gaps.
OpenAI explicitly uses AMD chips alongside Nvidia to diversify supply, and the overall compute shortage benefits all GPU/CPU suppliers that can fill gaps. Risk: AMD’s software ecosystem and datacenter GPU adoption remain behind Nvidia; execution risk on MI300/MI400 roadmaps.
AMD
Tae Kim Senior writer, Barron's; author of The Nvidia Way
Intel is named as a CPU supplier beneficiary in the author's take, and the overall compute demand explosion supports Intel’s datacenter CPU business and foundry aspirations.
Intel is named as a CPU supplier beneficiary in the author's take, and the overall compute demand explosion supports Intel’s datacenter CPU business and foundry aspirations. Risk: Intel continues to lose datacenter share to AMD and faces execution challenges in its foundry turnaround; product roadmaps are uncertain.
INTC
More from Key Context by Tae Kim
OpenAI Unveils GPT-5.5. Company Says Expect a Faster Model Release Pace
Apr 23, 18:10
Newswrap: TSMC Earnings, Nvidia I'm 'No Loser' CEO on Dwarkesh Podcast
Apr 16, 14:41
The Best Way to Invest in Memory Chip Stocks
Apr 14, 18:26
Exclusive: OpenAI Highlights Massive Compute Advantage Over Anthropic to Investors
Apr 09, 20:32
Newswrap: Amazon CEO Defiant on AI, Dell CEO Sees Memory Demand Explosion
Apr 09, 14:55

This newsletter, published April 24, 2026, features Tae Kim discussing MU, NVDA, AMD, INTC. 4 trade ideas extracted by AI with direction and confidence scoring.

Speakers: Tae Kim  · Tickers: MU, NVDA, AMD, INTC