Gray zone attacks: the crisis your comms plan was never built for.

Dear {{ first_name | reader }},

You've probably seen the phrase "gray zone" somewhere this year — next to Russia, a severed undersea cable, or a drone hovering over an airport nobody quite wanted to explain. It sounds like a term for people in uniforms, several floors above where crisis communication usually sits.

It isn't.

It started with a LinkedIn post. Consultant Aaron Marks has been writing about the gray zone regularly and reading his updates was enough to make me realise I couldn't have defined it properly myself. That's usually my cue to go and do the reading before I use a word like that in front of a client and hope nobody asks me to define it.

This week's edition works from a new Council on Foreign Relations playbook, Gray Zone Attacks Against U.S. Allies and Partners by F. David Diaz, read alongside the US intelligence community's own lexicon on the subject.

Both are written for a US policy audience, but the ambiguity they describe — deniability, delayed attribution, private-sector targets — doesn't stop at any one country's border.

The intelligence community's own assessment names China, Iran, North Korea and Russia as the states currently running these campaigns, and the case studies span the Baltic, East Asia and Latin America.

That ambiguity isn't a side effect. It's the entire design.

Enjoy!

Strip away the jargon and the definition is fairly plain. The intelligence community describes the gray zone as the space between ordinary diplomacy and outright armed conflict — a realm where states use coercive or subversive means, violent or not, to get what they want while staying just short of triggering a formal military response.

The Council on Foreign Relations puts it more bluntly: gray zone attacks are deliberate but deniable acts of violence, engineered to impose real costs while giving the attacking state room to deny involvement, at least at first.

The deniability is the point. A conventional attack forces a response, because everyone can see who did it. A gray zone attack is built so that the "who" stays unclear for as long as possible — long enough for the attacker to bank the advantage before anyone can act on it.

The current wave breaks down into four recognisable patterns.

Different methods, same shared flaw for the attacker to exploit: each one can plausibly pass as crime, accident, or coincidence, for as long as nobody looks too closely.

The CFR playbook calls this the campaign recognition problem, and it should worry a communicator more than the physical damage itself. It unfolds in two stages, one after the other, each one slow.

First, someone has to notice that a pattern exists at all. Gray zone incidents are spaced out and varied by design, so no single event looks like a campaign — a fire here, a cut cable there, a drone sighting somewhere else entirely.

Only in hindsight does the shape become visible. Second, once a pattern is finally recognised, someone still has to work out which state is behind it, which is its own separate, slower problem, because the whole point of routing an attack through a criminal proxy or an unregistered vessel is to make that step as hard as possible.

The 2023 Balticconnector incident is the case CFR keeps coming back to. A Chinese-flagged vessel dragged its anchor across a gas pipeline connecting Estonia and Finland, then slipped into Russian waters before anyone could act — not because nobody had a ship nearby, but because nobody had agreed in advance what to do.

That's a communications lesson as much as an operational one. Nobody lacked ships. What was missing was an agreed protocol for what to do next.

CFR is explicit that the primary targets of gray zone attacks abroad are frequently not government assets at all — they're commercial infrastructure, defence contractors, energy companies, logistics networks and private citizens connected to whichever state a campaign is aimed at, wherever in the world that happens to be.

That means the first phone call after an incident may land with a corporate comms director, not a foreign ministry spokesperson.

If your organisation touches energy, telecoms, ports, defence supply chains or any critical infrastructure with an international footprint, you're inside that target set, whether or not anyone has told you so.

I've long held that risk, crisis and emergency communication are three distinct disciplines, not three names for the same job.

Gray zone attacks are a good demonstration of why that distinction matters, because a single incident forces all three at once, and most crisis plans aren't built to move between them on the same afternoon.

The physical event demands emergency-style communication within the hour. The unresolved cause holds the situation in risk territory for weeks or months afterwards — probabilistic, hedged, genuinely uncertain.

And if attribution eventually lands, it can convert overnight into a full reputational and geopolitical crisis, one that retroactively reframes everything said in the two phases before it.

A plan built for one of those three moments will fail you in the other two.

Build these before the next incident, not during it. Separate your communications into two explicit tracks from the outset: what happened, and who is responsible.

The first track — facts, safety, operational status — can and should move fast.

The second should be held back and released only through coordinated, official channels, because premature attribution carries its own escalation risk. Draft holding language now for "cause under investigation" that doesn't quietly imply accident on one side or attack on the other; writing that sentence under pressure, for the first time, in front of a journalist, is a bad place to compose it.

Work out in advance who in your organisation actually talks to a host government, an embassy, or your national critical-infrastructure security authority — CISA in the US, and there's a comparable body in most jurisdictions — and build that relationship before you need it, not during the call.

CFR's own point about private-sector engagement applies directly here: treating that coordination as optional forfeits resilience you can't rebuild in the moment.

And don't let the absence of attribution become an excuse for silence on the facts your stakeholders are owed on day one — the two tracks move on different clocks, and only one of them can afford to wait.

There's no clean fix for the core tension here. CFR's own timeline gives attribution a window of 72 hours to 30 days, sometimes longer, and you will be under pressure — from journalists, from leadership, from your own instinct — to say more than you actually know well before that window closes.

Saying too much too soon has real cost. So does saying too little. This isn't a problem you solve with better wording. It's a discipline: keep the two tracks separate, be explicit about what remains unconfirmed, and update as the picture changes rather than trying to get it right in one statement.

Gray zone attacks aren't designed to be recognised as attacks — that's the whole mechanism. Which means the useful work for a communicator happens in the anticipate phase, long before the first fact needs stating: the protocols, the two-track language, the relationships with the people you'll need on day one.

By the time attribution is confirmed, the deniability has usually already done its work. Build for that now, and you're doing crisis communication.

Wait, and you're doing clean-up.

The bottleneck isn't what you want to say — it's how long it takes to type it. Wispr Flow removes the bottleneck.

Speak naturally and get polished, send-ready text for executive summaries, client updates, board recaps, investor notes, or just the 30 Slack messages you're behind on. Flow strips filler, formats numbers and lists, and preserves your tone.

Used by teams at OpenAI, Vercel, and Clay. 89% of messages sent with zero edits. Works in every app on Mac, Windows, and iPhone.

Start flowing free

_{Transparency & Disclosures}

_{AI Transparency: In alignment with EU AI Act requirements, please note that AI technology was used in the research, drafting, and/or image generation for this edition. All strategic analysis, professional opinions, and final editorial oversight are conducted exclusively by the author. Affiliate Disclosure: Some links in this briefing may be affiliate links. I only recommend tools and services I use personally or have vetted for professional efficacy. Professional Advice: This newsletter is for educational and informational purposes only and does not constitute legal or professional crisis management advice. © 2026 RiskComms FZCO. All rights reserved.}