Research

What happens after the announcement.

Long-form notes and short answers on the physical, commercial, and financial paths that determine whether energy and large-load projects can actually operate.

Archive

Long-form articles

13 articles
01
Geopolitics / strategic access

Why Venezuela and Greenland, now?

Venezuela has oil. Greenland has minerals and Arctic geography. The U.S. wants both, urgently, and for reasons that are starting to look like the same reason.

02
Refining / strategic routes

What makes Venezuela valuable now is not what it produces. It is the routes it does not have to take.

Hormuz, the diesel shortage, and Chevron's plans are giving another meaning to an industrial connection between Venezuela and the United States.

03
Field services / payment certainty

The payer is also part of the infrastructure

Venezuela can have a lot of work waiting and still have rigs sitting idle.

04
Physical execution / export system

Venezuela wants to produce more oil. The problem is how to get it out.

The Orinoco Belt needs diluent coming in through José before the crude can go out. The immediate problem is not a new port. It is recovering José—and figuring out who pays for it.

05
Energy security / system resilience

What energy security really means

Some interpret energy security simply as producing more oil. For others it is tied to renewables, protection from Russia, a larger strategic reserve, or less dependence on the Middle East.

06
Capital / payment architecture

Licenses don't pay the bills

Venezuela now has permits, barrel potential, and interested buyers. What still isn't defined is how the money actually gets to the oil projects.

07
Access / execution

The architecture now precedes the barrels

Washington announced majority control over 65 billion barrels. The fields, the companies and the contracts are still missing. Control on paper is moving faster than oil in the ground.

08
Data / field reality

The data problem nobody talks about when they talk about Venezuelan oil

Sometimes the map stops matching reality. Part of Venezuela's oil data may now have to be reconstructed from paper copies.

09
Cross-border gas / project execution

The easy project was the one that stopped.

Trinidad needs the gas to feed Atlantic LNG, sustain its petrochemical industry, and use infrastructure that has operated below capacity for years.

10
Energy systems / artificial intelligence

Where the narrative ends

Venezuela has the energy that artificial intelligence needs. That line felt inevitable—until the physical sequence became impossible to ignore.

11
Financial infrastructure / payment rails

Money needs a pipeline too

An energy system is incomplete if money cannot move legally, verifiably, and predictably between the parties—even when the gas, turbines, and buyers are already there.

12
Gas / competing demand

Who ends up with the gas?

Trinidad has something Venezuela still needs to build: infrastructure that can receive gas, process it, sell it, and turn it into commercial cash flow.

13
Offshore gas / phased recovery

BP, Loran, and why Venezuela won't recover the way you think

BP received the Phase 2 license for Venezuela's Loran offshore gas field alongside XRG and UCC, with each company holding an equal stake.

Compute archive

Short Answers

Concise notes on power, interconnection, equipment, water, and large-load deployment.

20 answers
01

How is AI infrastructure changing Big Tech capital intensity?

AI Infrastructure Capex Shift

02

Why can skilled labor become a constraint on AI infrastructure deployment?

Labor Constraints in AI Infrastructure

03

Why can capacity-market signals move faster than physical infrastructure?

Capacity Markets vs Physical Infrastructure Timing

04

Why is AI infrastructure moving beyond GPU-only systems?

Compute Architecture Fragmentation

05

Why could compute capacity become an institutional infrastructure asset class?

Compute Capacity as an Asset Class

06

What is constraint intelligence and why does it matter for infrastructure deployment?

Constraint Intelligence for Infrastructure

07

How do interconnection delays affect AI data center deployment?

Interconnection Delay and AI Infrastructure

08

Why can non-firm power create risk for high-uptime compute loads?

Non-Firm Power and AI Data Centers

09

Why are nuclear long-lead components entering AI-era power planning?

Nuclear Supply Chains and AI Power Planning

10

Why are AI factory campuses moving toward integrated on-site power models?

On-Site Power for AI Factory Campuses

11

Why is PJM becoming a constraint node for AI infrastructure?

PJM Compute Constraint Stack

12

Why is available power not the same as deployable infrastructure?

Power Availability vs Deployability

13

How are AI data centers increasing dependence on power equipment supply chains?

Power Equipment Supply Chains and AI Data Centers

14

Why is second-life battery storage relevant to grid-constrained power demand?

Second-Life Battery Storage and Grid Constraints

15

Why are power transformers becoming a bottleneck for AI infrastructure?

Transformer Supply and AI Grid Expansion

16

Why is AI infrastructure increasing energy demand so rapidly?

AI Data Center Energy Demand Growth

17

Is electricity the main constraint for AI infrastructure deployment?

Electricity as the Primary Constraint in AI Infrastructure

18

How do cooling and water affect AI infrastructure deployment?

Cooling and Water Constraints in AI Infrastructure

19

What determines where data centers can be built?

Data Center Site Selection Constraints

20

Can power grids keep up with AI infrastructure growth?

Grid Limitations vs AI Growth

Follow what changes the operating decision.

Get new Frontier Grid notes by email.

Get operating updates
Energy & Infrastructure Research | Frontier Grid