Two things happened to energy companies in the last eighteen months, and most leadership teams are only paying attention to one of them, if either.
The first: your buyers, partners, investors, and prospective hires stopped Googling you. They started asking. When someone wants to know whether your company is a credible operator, a safe JV partner, or a place worth taking a job, a growing share of them now type the question into ChatGPT, Perplexity, or Google’s AI answer box, and they read whatever the machine says back. They rarely click through to your website to check. The AI’s summary of your company is your company, as far as that first impression goes.
The second: AI quietly moved inside your operation. It’s drafting reserves narratives, flagging equipment for maintenance, informing trading positions, screening résumés, and sitting one query away from your HSE and capital-allocation conversations. Most of it arrived without a policy, without an audit trail, and without anyone able to answer the question a regulator or an investor will eventually ask: who decided this, on what basis, and can you show me?
These are the two fronts. Outward, AI is deciding how the market sees you. Inward, AI is deciding things for you. Energy is more exposed than almost any other industry on both. Long sales cycles, heavy due diligence, real regulatory weight, and capital that moves on reputation. Here’s why each one matters, and what to actually do.
Front One: AI Is Describing Your Company to Buyers Right Now
Run the test yourself. Open ChatGPT or Perplexity and ask, “Is [your company] a reliable midstream operator?” or “Who are the top operators in [your basin]?” Read the answer as if you were a procurement lead, a capital partner, or an engineer weighing an offer.
If the answer is thin, outdated, generic, or simply omits you while naming three competitors, that’s not a marketing inconvenience. In a business where a single JV or supply agreement runs into the millions, it’s a missed shortlist. And it compounds: AI systems cite the sources they can read clearly, so the companies that show up today get cited more tomorrow.
The fix is not “do more SEO.” It’s making your company legible to machines: a clean, consistent description of who you are and what you do, structured so an AI can extract it; third-party presence on the credible sources these systems actually pull from; and content that answers the specific questions buyers ask, in plain language. This discipline now has a name, answer engine optimization, and the research behind it is no longer speculative. (It’s the work my agency, EWR Digital, spends most of its time on for energy and industrial clients. But the principle stands whoever does it: if the machines can’t read you, the buyers asking the machines won’t find you.)
Front Two: AI Is Making Decisions Inside Your Company Without a Paper Trail
The inward front is quieter and, frankly, more dangerous. Ask your own teams where AI already touches a decision. You’ll find more than you expect, and almost none of it documented.
The exposure isn’t hypothetical. Reserves and production figures feed disclosures. Trading models move real money. Maintenance and HSE calls carry safety and liability weight. Hiring screens carry discrimination risk. The moment an AI tool informs one of those, you’ve created a decision that, today, you probably can’t reconstruct. When an auditor, a regulator, a plaintiff’s attorney, or an institutional investor asks “how was this determined and what controls were in place,” “the model suggested it” is not an answer that survives the room.
What governs this isn’t more technology. It’s a framework, the kind formalized in the NIST AI Risk Management Framework: an inventory of where AI touches material decisions, a risk tier for each, a human accountable for each tier, and a record that lets you reconstruct any decision after the fact. Boards already do this for financial controls and HSE. AI decisions now need the same treatment, and the energy companies getting ahead of it are treating it as a governance question, not an IT one. (This is the problem my other company, ModalPoint, was built to solve. Think of it as controls and audit trails for the decisions your AI is now involved in. The operators getting this right are building that record now, before someone outside the company asks to see it.)
Same Root Cause, Same People Who Win
Pull back, and the two fronts are the same story: AI moved faster than your governance and faster than your digital presence, and the gap is now visible to the outside world. One gap shows up as buyers who can’t find you. The other shows up as decisions you can’t defend. Both land on the same desk, yours.
The leaders pulling ahead aren’t the ones who bought the most AI tools. They’re the ones who asked two boardroom questions early:
| Front | The question to ask | Who owns it |
|---|---|---|
| Outward | When a buyer asks an AI about us, is the answer accurate, current, and present? | Marketing / commercial |
| Inward | Where is AI already shaping our material decisions, and can we reconstruct them? | Risk / governance / the board |
Neither question requires a moonshot. Both require someone senior to own it before the market, or a regulator, forces the issue. In oil and gas, where reputation and defensibility have always been the currency, the companies that answer these two questions first will quietly take ground from the ones still arguing about which chatbot to license.
The two-front problem is already here. The only choice left is whether you’re managing it or discovering it.
Matthew Bertram is involved in AI visibility, digital transformation, and industrial AI governance initiatives across the energy sector through OGGN, ModalPoint, and EWR Digital.