This story is currently available in Αγγλικά only — a translation isn't ready yet.
When a country loses power nationwide three times in two weeks, the failure has stopped being technical. Someone in Tbilisi has made a choice about which risks to price.
ℹ️ Ανάγνωση από τον περιηγητή · στούντιο φωνή AI έρχεται σύντομα

A power grid that collapses once is an accident. Twice is a warning. Three times in fifteen days is a policy — someone has decided that the cost of preventing the next blackout is higher than the cost of absorbing it, and that someone works for Irakli Kobakhidze.
On the evening of August 5, Georgia went dark again. It was the third nationwide outage since late July. The Interior Ministry issued its usual reassurances; the transmission operator will, in due course, produce a report with the word "cascading" in it. But the interesting question has nothing to do with the substation that tripped. The interesting question is why a European candidate country with a functioning economy, a Black Sea coast, and two decades of donor investment in critical infrastructure now switches off with the regularity of a Soviet suburb in February.
Georgia's grid was rebuilt largely on external money after 2003. The hardware is not the problem. The problem is what the hardware has been asked to do over the last three years, and by whom. Since the ruling party pivoted its economic model toward absorbing whatever Russian capital, tourism and re-export flow arrived after 2022, electricity demand has done things the network was never dimensioned for. Data centres for crypto operators pushed out of Kazakhstan. Aluminium and cement pulled from sanctioned supply chains. Summer tourist load in Batumi that now behaves like an industrial customer. The state's own numbers, when it publishes them, show consumption growing faster than transmission upgrades.
This was a bet. The bet was that the arbitrage — sanctions arbitrage, re-export arbitrage, energy arbitrage — would generate enough revenue to pay for the grid it was overloading, and that the bill would arrive after the next election rather than before it. The bet has now been called, on a Tuesday evening, in the dark.
What makes the third blackout politically expensive rather than merely embarrassing is what happened in the same news cycle. László Róbert Mézes, a Hungarian journalist who has covered the anti-government protests for years from inside Tbilisi, was picked up by migration officers on the same day the lights went out. The United National Movement, gathering in congress, elected an interim governing council under Irakli Pavlenishvili and produced the sort of communiqué that oppositions produce when they sense the ground is finally moving. None of these three things caused the others. All three are legible as the same story: a government that has decided the friction of running the country is now higher than the friction of narrowing it.
The muscle memory here goes back further than the current cabinet. Georgian governments have, since independence, tended to treat infrastructure as a political asset — something to be delivered visibly before a vote and left to depreciate quietly after one. What has changed under Kobakhidze is the composition of the demand curve. The customers whose consumption is now stressing the transmission network are, disproportionately, the customers whose presence in Georgia is the ruling party's economic thesis. You cannot lecture a Russian-owned data centre about load-shedding when its arrival was, six months ago, filed under "foreign direct investment."
The European framing of this is where most commentary goes wrong. Brussels reads Georgian blackouts as a governance signal — evidence that candidate status was granted prematurely, that the rule-of-law backsliding has an operational cost, that the country is drifting. This is true and largely useless. Nothing in the Commission's toolkit fixes a substation. The leverage that would matter — conditioning the remaining tranches of macro-financial assistance on transparent grid audits, for instance — requires a political appetite in Brussels that the current Commission has already shown it lacks when the target is a government it has publicly written off.
Moscow reads the same blackouts differently, and this is where the analysis has to be honest. A Georgia whose grid is fragile, whose journalists get detained, whose opposition is regrouping in a congress hall rather than on the street, is a Georgia that has stopped being a demonstration project for anyone. That suits the Kremlin. It costs nothing. It requires no tanks. The Russian budget line for "destabilising Georgia" in 2026 is, one suspects, remarkably small, because the work is being done by the electricity bill.
What would prove this reading wrong: a serious, published, independent audit of the July–August outages, with named responsible parties and a costed remediation plan, delivered before the autumn. What would confirm it: another blackout, another vague statement, and a quiet reshuffle at the transmission operator in which no minister loses their job.
I would price the second scenario at roughly four times the first. The grid, unlike the government, does not lie about what it can carry.
Georgia has suffered three nationwide blackouts in two weeks—not a technical failure but a policy choice by Prime Minister Irakli Kobakhidze's government, which has bet that arbitrage profits from Russian capital and sanctions-evasion would pay for grid upgrades before the bill came due. The outages coincided with the detention of a Hungarian journalist and opposition regrouping, signalling a government prioritising narrower control over functional state infrastructure.
If Georgia's grid failures continue unchecked, the country risks losing what remains of its appeal to foreign investors and EU partners—a cascade that could accelerate capital flight and regional instability. For anyone doing business in Georgia or tracking Caucasus geopolitics, the next blackout will signal whether the government intends to fix the problem or whether it has chosen managed decline as preferable to losing control.