The Perception Gap Index hit 5.88 on March 28, with 42 stories scanned across seven regions. The geopolitics tributary crossed into Competing Realities at 7.02 — the only stream above 7.0 today. The widest regional pair: US ↔ Middle East at 7.69. The day's top story, Israel's renewed nuclear strikes on Iran, scored 8.13. The same military action is four different wars depending on which language you read it in.
That number, by itself, is just a number. Here's what it means.
The same bombs. Four different wars.
Israel struck Iranian nuclear sites at Natanz and Arak again on March 28. AP led with Israel killing an Iranian commander and Trump saying "keep blowing them away." The Guardian led with Iran's retaliation — "Iran launched a barrage of missiles targeting Israeli nuclear sites." Al Jazeera filed it under "war enters fifth week." India's NDTV focused on "Iran missile strike on Dimona stress tests Israel's nuclear ambiguity."
Four entry points to the same event. Four different wars.
In Washington, this is a justified military campaign with diplomatic leverage. The [nuclear strikes](/blog/iran-dimona-natanz-nuclear-exchange-perception-gap-2026) are pressure applied correctly. In Brussels, it's a dangerous escalation spiral — both sides retaliating against nuclear facilities in a pattern no one controls. In Tehran and Doha, it's the "American-Israeli war on Iran" — that phrase, repeated verbatim across Arabic-language outlets, turns a bilateral conflict into a civilisational one. In New Delhi, it's a case study: The Hindu compared it to Israel's 1981 Osirak reactor strike in Iraq, treating the present as footnote to a longer argument about whether nuclear deterrence still works.
The PGI of 8.13 comes not from anyone lying. Every fact checks out. The divergence lives in which facts lead, which actors initiate, and which framework surrounds the data. AP says "pressuring Iran simultaneously through military action and diplomatic deadlines." The Guardian says "retaliatory strikes marked a dangerous escalation in the tit-for-tat nuclear targeting." Al Jazeera says "American-Israeli war on Iran." Same day. Same bombs. Near-opposite stories.
When the wounded become invisible
The day's second-highest PGI — 7.53 — belongs to a story most Americans barely registered: more US troops wounded at a Saudi base.
US outlets framed it as a support-the-troops moment. Arabic-language outlets framed it as foreign occupation of sovereign and sacred soil. The factual gap mattered here — Arabic sources provided more granular damage reports, including specific facility hits and casualty details absent from English coverage. The US ↔ Middle East pair hit 8.5 on this story alone.
But the real perception gap isn't between regions. It's between this story's significance and its visibility. American soldiers are being wounded in a Middle Eastern country. In previous conflicts, this would dominate domestic news cycles. In March 2026, it's a paragraph below the market numbers.
The [Houthi strike on Beersheba](/blog/iran-gulf-states-missiles-trapped-between-us-perception-gap-divided-march-2026) tells a parallel story. PGI 7.30. The US labels Houthis as Iran-proxy terrorists. The Middle East frames them as a resistance movement. The EU focuses on escalation risk. Three framings, none of which communicate with each other.
The River System: where the fractures run
The seven tributaries of today's information river tell a clear story about where the world's narrative fissures concentrate.
The red zone. Geopolitics (PGI-GP) at 7.02 — Competing Realities. This is the only tributary above 7.0. Fourteen stories fed this stream today, and the top five all involve the Iran conflict from different angles: nuclear strikes, wounded troops, Houthi attacks, Rubio's G7 diplomacy, and Trump's deadline extension. The geopolitics stream isn't just running hot. It's boiling.
The orange zone. Five tributaries sit in Diverging Narratives territory. Economics (PGI-EC) at 5.86, driven by the [Iran war's $300 million daily cost](/blog/iran-war-costing-300m-daily) and diverging interpretations of the Dow's correction. Information Warfare (PGI-IW) at 5.79, pulled up by the NeurIPS China ban reversal — a story that's either a routine policy correction or evidence that Beijing can resist Western sanctions overreach, depending on your continent. Climate/Energy (PGI-CL) at 5.31 with eleven stories, the broadest tributary by volume. Technology (PGI-TE) at 5.13. Women's Rights (PGI-WR) at 4.66.
The calmest stream. Health (PGI-HE) at 4.58. Three stories, limited coverage, modest divergence. When Sudan's health system being 80% destroyed barely registers on a global perception index, the problem isn't low divergence. It's low visibility. The world largely agrees about the health catastrophe — because the world largely isn't looking.
The pattern: the closer a story sits to the Iran war's blast radius, the wider the perception gap. Nuclear strikes hit 8.13. Wounded troops hit 7.53. Houthi strikes hit 7.30. But move to the conflict's second-order effects — economic damage, energy transition, health collapse — and the PGI drops below 6.0. Not because those stories matter less. Because fewer regions cover them, and fewer regions means fewer opportunities for narrative disagreement.
Low PGI doesn't always mean the world agrees. Sometimes it means the world isn't watching.
Cui bono: who profits from each version
Every region's framing of the Iran war serves identifiable interests. This isn't conspiracy — it's economics. Narratives, like markets, are shaped by the interests of their producers.
The nuclear strikes. US coverage frames Israel's Natanz and Arak strikes as justified deterrence. This framing serves the administration's "escalation is the path to peace" narrative and protects ongoing military appropriations. EU coverage frames the same strikes as a dangerous escalation spiral, which serves European diplomatic positioning — Brussels can present itself as the adult in the room calling for ceasefire. Middle Eastern coverage frames it as the "American-Israeli war on Iran," which serves regional solidarity and justifies Iran's retaliatory posture. South Asian coverage frames it as a nuclear norms crisis, which protects India's careful relationship with both Washington and Tehran while defending the deterrence framework on which India's own nuclear posture depends.
The deadline extension. Trump's extension of the Hormuz deadline to April 6 becomes three different stories with three different beneficiaries. The New York Times framed it as a market-responsive move — S&P down 1.7%, Brent up 5.7% to $108 — which serves Wall Street's narrative that financial markets moderate foreign policy. The Guardian framed it as contradictory strongman politics — Trump "continued to declare victory" while Iran "strenuously denied" it was begging for a deal — which serves European calls for genuine diplomacy over theatre. Al Jazeera framed it as a pause in potential war crimes, citing Geneva Convention experts — which serves the legal case against US military action.
The oil windfall. Identical [Big Oil profits](/blog/big-oil-63-billion-iran-war-windfall-gas-prices-2026) produce opposite conclusions across the Atlantic. The New York Times presents oil executives as conflicted — "It feels like you should be jumping up and down. But there is a picture out there where the economic realities get very, very challenged." This serves the industry's PR by making windfall profits sound worrying rather than celebratory. The Guardian frames the same profits as proof the fossil system is broken — windfall money "flows back into exploration, extraction and export infrastructure," creating a vicious cycle. This serves the EU's green industrial policy and the argument that renewables aren't just ethical but economical.
The interest-alignment pattern is consistent: each region's dominant framing of the Iran war serves its own institutions, economic structures, and geopolitical positioning. Nobody is lying. Everybody is selecting.
The economic war you're reading about depends on your currency
The war's economic damage is real and global. But the explanation for that damage shifts depending on where the journalist sits.
The New York Times covers Asian currencies "crumbling as governments race to secure fuel priced in American money." The causal mechanism is dollar hegemony — the structure of the global financial system. The Guardian covers the same economic pain but attributes it directly to "US-Israeli attacks on Iran." The Guardian Australia adds that G20 inflation will hit 4% through 2026, "1.2 percentage points higher" than forecast.
One blames economic architecture. The other blames military action. They can't both be the primary cause.
The NYT framing inadvertently highlights American structural advantage — dollar hegemony means the US benefits from oil being dollar-denominated even as others suffer. The Guardian framing names the war itself as the culprit, assigning agency and responsibility. For an Australian reader, the message is clear: your economy is hurting because of a war you didn't start. For an American reader, the message is different: the system is under strain, but the system is ours.
Putin's request for oligarch war funding shows a similar split even within the Western press. Reuters gave the Kremlin denial equal weight — "he said, she said." The Guardian treated it as established fact, evidence sanctions are crippling Russia. The Financial Times read the same report as proof Putin will fight on regardless of cost. Same leak. Three interpretations. Two of them support continued Ukraine aid, but for contradictory reasons — one says Russia is breaking, the other says Russia won't stop.
The AI cold war you didn't notice
Buried below the nuclear headlines, a quieter perception gap played out in a conference room.
NeurIPS — the world's most prestigious AI conference — reversed its ban on papers from Chinese sanctioned entities. Reuters reported a straightforward policy correction: the Chinese boycott "forced the reversal." The South China Morning Post reported a sovereignty victory: NeurIPS "apologises" and China successfully defended "the core values of academic exchange."
Same reversal. One region sees a conference fixing a mistake. The other sees a superpower successfully resisting Western sanctions overreach through coordinated institutional pressure.
The PGI was 5.1 — modest for a bilateral story. But the Cui Bono dimension scored 5.5, because the interest-alignment is sharp. Reuters' framing serves Western academic institutions that want to maintain China ties. SCMP's framing serves Beijing's narrative that coordinated pushback works — and that China can maintain "access to cutting-edge Western AI research while building its own parallel ecosystem."
This story scored lower than the nuclear strikes. It matters more for the decade ahead.
Global Attention Index: what the world can't see
Today's GAI — the Global Attention Index, measuring which stories reach which populations — hit 5.33: Selective Visibility. The world isn't blind. But it's looking through a keyhole.
The attention desert. Health (GAI-HE) is the only tributary in the Information Shadow tier at 6.30. A COVID variant called Cicada has reached 23 countries. It got coverage in exactly one region: the United States. A hyper-mutated variant spreading while health systems are stretched by war logistics, and 94% of humanity can't see the story. Sudan's health system — 80% destroyed, 519 facilities attacked — is visible only in Africa.
The global spotlight. Economics (GAI-EC) at 4.43 is the most visible tributary. The fertilizer crisis — plants closing across three continents, the UN building a wartime corridor just to move fertilizer through Hormuz — broke through to five regions. When the UN institutionalises a crisis, the crisis becomes visible. Attention follows institutions, not suffering.
The invisible children. Somalia's hungriest children — "too weak to even cry," in UNICEF's words — scored a GAI of 6.86, the day's most invisible story. Seen by Africa and a buried US mention. Invisible to the EU, Middle East, South Asia, Asia-Pacific, and Latin America. Approximately 5.7 billion people can't see this story. It scored a PGI of just 1.8 — rare global consensus. Where it's covered, everyone agrees: these children are dying because of the war's cascading effects on fuel, food, and humanitarian delivery. The perception gap isn't in the framing. It's in the seeing.
The PGI × GAI crossover. The most instructive comparison today: Israel's nuclear strikes scored PGI 8.13 and GAI 3.15 — the world sees it, and sees it very differently. Somalia's hunger crisis scored PGI 1.8 and GAI 6.86 — where it's covered, everyone agrees, but most of the world doesn't see it at all. These are opposite pathologies of the global information system. One produces conflicting realities. The other produces shared blindness.
Region blindness. Latin America missed 95.2% of today's 42 stories — an entire continent of 660 million people seeing almost nothing. Africa missed 83.3%. The US and EU each missed 26.2%. The information system isn't just unequal. It's mutually blind. The periphery can't see the centre, and the centre won't see the periphery.
The consensus that no one noticed
Amid all this divergence, one story achieved something rare: near-total agreement across regions.
UNICEF's Somalia report — children starving because the Iran war's fuel price cascade made humanitarian deliveries impossible — scored a PGI of 1.8. Global Consensus. Both AP and the Washington Post reported from the same Dollow displacement camp, cited the same UNICEF executive director, used the same devastating details. "Children too weak to even cry." "Mothers struggle to keep their children alive." Same causal framework: distant war amplifying local drought. Same emotional register: designed to break through compassion fatigue.
The WaPo pulled one thread harder — connecting the dots for US readers between military action and humanitarian consequences. But the divergence was minimal. Where this story was seen, the world agreed.
The problem is the where. Two regions. Two outlets. A consensus nobody heard.
What 5.88 tells us
A PGI of 5.88 sits in the upper range of Diverging Narratives, pulled there by a geopolitics stream in Competing Realities. The number means the average story today produced meaningfully different narratives across regions — not different enough for alternate realities on most topics, but for anything touching the Iran war directly, the world is watching different movies.
The ME ↔ US pair at 7.69 is the day's defining fracture. Thirteen stories separate these two regions' information ecosystems, and on every one, the framing differences are structural, not incidental. It's not that American and Middle Eastern journalists disagree about interpretations. They disagree about what the story is.
The AP ↔ EU pair at 4.93 is the day's closest alignment — Asia-Pacific and European outlets share enough framework to be reading recognisably the same story, even when they emphasise different elements. Japan's oil reserve release is either Tokyo leading or Washington leading, but at least both sides agree it's a coordinated response to a supply crisis.
Between these two poles — 7.69 and 4.93 — sits the day's information reality. Not a single global conversation. Not entirely fractured either. A world where some pairs of regions can talk to each other and others can't.
The energy transition debate captures this perfectly. The same [oil crisis](/blog/hormuz-crisis-accelerating-clean-energy-transition-asia-2026) is simultaneously evidence that the green transition is "dangerously messy" (US) and that renewables-heavy economies are proving resilient while fossil-dependent ones suffer (EU). The NYT gives "fossil loyalists" equal airtime. The Guardian and BBC foreground Spain and Portugal's declining electricity prices, Octopus Energy's 50% surge in solar panel sales. The US says "it's complicated." Europe says "renewables are working." Neither is wrong. Together, they're a perception gap you could drive a wind turbine through.
One number. Forty-two stories. Seven regions. Four wars inside one war. And the children too weak to cry, invisible to 5.7 billion people who agree — when they can see them — that something has to change.
The question worth sitting with: if you read the same nuclear strike through four different moral frameworks today, which one felt like "the news"? That's the one doing the most work on your perception.