Windows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallCrashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minuteiTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
Apollo Global Management President Jim Zelter reportedly identified compute and energy as potential constraints on AI’s expansion. That is an executive view, not a measured forecast: the available October 6, 2026 Bloomberg synopsis does not include the full interview or its detailed reasoning. Independent International Energy Agency data show why the concern matters, while also making clear that the scale and timing of any constraint vary by location and remain uncertain.
What did Jim Zelter say about AI bottlenecks?
A short synopsis of a Bloomberg item dated October 6, 2026, reports that Zelter discussed compute and energy bottlenecks as AI investment accelerates. It also says he discussed Apollo’s AI and data-center exposure, the U.S. economy, private capital in defense, and European investment. The full Bloomberg story or transcript is not available in the cited material, so it does not establish Zelter’s detailed arguments, figures, or specific portfolio positions.
The claim is best understood as a warning about potential constraints, not proof that AI growth will inevitably stall. Apollo’s June 2026 mid-year outlook separately framed the issue around compute demand, compute price, and compute supply, and posed the question of whether energy and data-center capacity can keep pace. The discussion asks useful questions; it does not, by itself, quantify a future shortage. Apollo’s mid-year outlook
What the electricity figures show—and what they do not
The International Energy Agency (IEA) estimated that data centers consumed about 415 terawatt-hours (TWh) of electricity worldwide in 2024, approximately 1.5% of global electricity use. In its 2025 report, the IEA’s base case projected data-center consumption of about 945 TWh in 2030. That is a scenario, not a guaranteed outcome; the agency discusses uncertainty and uses multiple scenarios. Neither global figure shows how much capacity is available at a particular site or whether a given region faces a shortage. IEA, Energy and AI
#1 Best Overall
- Brand New, High Quality Replacement Cord
- Tested Units. In Great Working Condition.
The IEA reported in 2026 that data-center electricity demand grew 17% in 2025, faster than overall global electricity demand, with demand from AI-focused data centers rising faster still. This is evidence of rapid growth, not evidence that every electricity grid is under equal pressure. Local impacts depend on where facilities are built and on the generation, transmission, and grid capacity available there. IEA, 2026 data-center electricity update
Where a bottleneck could arise
“Compute” and “energy” are connected, but they are not single constraints. AI growth depends on obtaining equipment, building facilities, and supplying reliable power at the right place and time. A global increase in electricity use cannot identify which link is limiting a particular project.
| Potential constraint | What it means |
|---|---|
| Compute supply | Accelerators and servers must be available and installed; having equipment alone does not ensure that it can be operated at scale. |
| Power generation | Enough electricity must be generated to serve new data-center demand. |
| Grid delivery and connection | Transmission and local grid capacity must deliver power to the facility, and a project must secure a grid connection. |
| Construction and equipment lead times | Data-center construction and delivery of components can take time; the IEA identifies supply-chain tightness for items such as transformers and gas turbines. |
| Planning and approvals | Permitting, planning, and regulatory processes can affect how quickly facilities and supporting infrastructure are built. |
| Workload growth and efficiency | Electricity use per AI task can fall as hardware and software improve, while wider adoption or more energy-intensive uses can still raise total demand. |
The IEA’s 2026 update points to supply-chain tightness, delayed grid connections, and approvals as practical constraints on near-term expansion. In other words, adding generation alone may not resolve a local constraint if a project cannot connect to the grid, obtain components, or complete construction in time. IEA, 2026 data-center electricity update
Why AI’s power needs are not just about total consumption
AI servers have become more power-dense, and the IEA says rack power density is expected to rise further by 2027. AI training and use can also involve large, rapid swings in power consumption. These characteristics make reliable delivery and, potentially, storage relevant alongside the total amount of electricity a data center uses. A projected or peak rack demand should not be treated as typical consumption for every rack. IEA, Energy and AI executive summary
Rank #3
Efficiency complicates the outlook. The IEA says electricity use per AI task is declining rapidly, but greater use of AI and more energy-intensive applications can push aggregate consumption upward. Lower energy use for an individual task therefore does not, on its own, settle whether total demand will rise or fall. IEA, 2026 data-center electricity update
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.How to interpret the bottleneck claim
- As a risk to monitor: demand growth and infrastructure constraints make compute availability and reliable power relevant to the pace of AI deployment.
- Not as a certain prediction: the reported Bloomberg synopsis does not provide Zelter’s full reasoning, and the IEA’s 2030 figure is a base-case projection rather than a promise.
- As a local question: grid connections, available capacity, permitting, and equipment supply can matter more to an individual project than a global demand share.
- Alongside efficiency: falling power use per task may temper demand, but increased use can offset those savings.
The practical question is not simply whether the world has enough electricity in aggregate. It is whether compute equipment, data centers, grid connections, and dependable power can arrive in the right places quickly enough to serve growing AI workloads.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
