President Trump said Friday he was bringing his national security team into the Situation Room for a final decision on a tentative U.S.-Iran agreement, turning the most consequential live war story into a test of signature power.

The reported deal is not peace. It is smaller, more useful, and more fragile than that: a 60-day memorandum of understanding that would extend the ceasefire, open another round of talks on Iran's nuclear program, and try to keep the Strait of Hormuz from becoming a tollbooth with missiles. AP reported that negotiators had reached tentative terms. Axios reported that U.S. officials described a draft requiring unrestricted shipping through Hormuz, no tolls, no harassment, and removal of mines from the strait within 30 days.

That is the narrow hinge on which a large amount now turns. The war has already taught the region that declarations and conditions are not the same object. On Monday, the U.S. military said it carried out self-defense strikes in southern Iran, including missile launch sites and boats laying mines, even as the administration said negotiations were moving. That is the ceasefire in miniature: armed restraint, reversible language, and enough firepower near the page to remind everyone that paper can burn.

The politics in Washington are no less evasive. On May 1, Trump notified Congress that hostilities the United States began against Iran on Feb. 28 had been terminated, while leaving open future military action. The Senate had blocked a war-powers resolution meant to force authorization or an end to military action. The administration's theory has been plain enough: if the fighting can be defined as episodic, then the clock can be managed. If the clock can be managed, Congress can be made ornamental.

That makes today's meeting more than another White House drama. It is the place where three forms of leverage collide. Iran wants relief from the blockade pressure and the risk of renewed strikes. Global shipping wants a channel, not a promise. Trump wants an agreement that can be sold as both victory and restraint without conceding that Congress had any right to supervise the path there.

The hazard is not only that Trump rejects the draft. It is that he accepts it as a prop. A 60-day ceasefire extension can be a bridge to a nuclear bargain, but it can also be a cooling-off room before the next round of escalation. The language about Hormuz matters because it is measurable. Are mines removed? Are tankers harassed? Are insurers reassured? Does oil move because a state changed behavior or because traders decided to believe a headline for a few days?

The better question for the public is not whether this is peace. It is whether anyone has built an enforcement mechanism strong enough to survive the first useful violation. Deals that depend on personal approval can arrive quickly and vanish the same way. The Strait of Hormuz can be reopened on paper before it is reopened in risk models, naval posture, or the minds of ship captains.

Still, the table matters. The hour's difference is that negotiators appear to have produced something specific enough for a president to approve or kill. In a war that has repeatedly been narrated as inevitability, the existence of a draft is an accusation. Someone now has to choose.

The Gaza peace architecture now has enough titles to fill a summit badge and not enough authority to move rubble.

AP reported Thursday that plans for an International Stabilization Force for Gaza are in doubt as troop pledges stall. The force was announced in February with ambitions of reaching 20,000 personnel under the umbrella of President Trump's Board of Peace. The promise was order after catastrophe: security, disarmament, reconstruction, a governing transition. The operating reality is a series of locks with different owners.

One lock is military. AP reported earlier this month that Nickolay Mladenov, the envoy overseeing the U.S.-brokered ceasefire, acknowledged the truce was stalled over Hamas disarmament, paralyzing reconstruction. Israel has continued strikes, Hamas has not given up weapons, and the map left by the ceasefire still divides control rather than settling it.

Another lock is financial. Reuters reported that the Board of Peace warned of a funding gap in a reconstruction plan estimated at $70 billion. Money pledged at diplomatic volume is not the same as money disbursed at construction speed. Gaza needs cement, policing, sewage, authority, and a calendar. It is receiving conditions.

A third lock is political legitimacy. Le Monde reported that the Palestinian committee meant to help administer Gaza remains effectively frozen, with members describing the need for a stable security climate, one authority, and reliable financing before it can function. That is a reasonable list. It is also a description of the very outcome the committee was supposed to help create.

This is the trap of postwar governance by brand. The phrase Board of Peace suggests a command center. The facts describe a waiting room. The board cannot rebuild Gaza without demilitarization. It cannot get demilitarization without incentives and guarantees. It cannot supply incentives without money and political trust. It cannot generate trust while people on the ground see the ceasefire as conditional, partial, and perhaps temporary.

The result is a peace plan that can still be defended in speeches but not yet experienced as governance. For Palestinians in Gaza, that distinction is not academic. A committee that cannot govern is not transitional authority. A stabilization force without troops is not security. A reconstruction fund without cash is not reconstruction.

The war did not end because a new institutional vocabulary arrived. It paused in pieces. Gaza is now being asked to live inside the difference.

The most important immigration story in the United States may be architectural: fewer forums, narrower speech, less time, less bond, more removal.

The Supreme Court sided this week with the Trump administration in a dispute over speech restrictions for immigration judges, a case that touched the rights of federal workers who occupy one of the strangest roles in American law. Immigration judges are called judges, but they sit inside the executive branch. Their independence has always been less secure than the word on the door suggests.

That door is closing in some places. AP reported that San Francisco's main immigration court shut down May 1 after a purge of judges left it with only two judges, down from 21 when Trump returned to office last year. A court does not need to be formally abolished to stop functioning. It can be reduced until the docket becomes its own denial.

The detention side is moving just as aggressively. AP has documented how a no-bond approach for immigrants in custody, long used by a group of judges in Tacoma, became a preview of a broader administration policy now moving toward a likely Supreme Court fight. Human Rights Watch, in a report covered by AP, said the administration had deported nearly 13,000 Cubans, Venezuelans and other nationals to Mexico, where many face danger in a country not their own.

The administration frames the effort as enforcement finally taken seriously. Its critics call it due process turned into a maze. The more precise description is institutional compression. The government is shrinking the number of meaningful moments at which a person can be heard.

That compression has political advantages. It is harder to campaign against a hearing that never happens than against one dramatic abuse. It is harder to explain why judicial speech rules matter than to argue about a border crossing. It is harder to photograph a docket backlog than a raid.

But the system's legitimacy lives in those unglamorous parts. Bond hearings, judge independence, courthouse access, and appeal windows are not procedural decorations. They are the machinery that distinguishes enforcement from administrative force.

A country can choose a stricter immigration policy. It should not pretend that disabling the referee is the same as winning the argument.

The U.S. economy is not giving policymakers the courtesy of a clean story.

The Bureau of Economic Analysis said Thursday that the PCE price index rose 3.8 percent in April from a year earlier. Core PCE, excluding food and energy, rose 3.3 percent. Both sit well above the Federal Reserve's 2 percent target. At the same time, the BEA revised first-quarter real GDP growth down to a 1.6 percent annual rate from its earlier 2.0 percent estimate.

That combination is unpleasant but clarifying. The problem is not that the economy has stopped. It is that the part the Fed wants softer, inflation, remains stubborn while the part it would like sturdier, underlying growth, looks less impressive on revision.

Consumer behavior is doing its usual work of confusing everyone paid to predict it. Personal consumption expenditures rose 0.5 percent in April, while disposable personal income slipped 0.1 percent. In plain English: households spent more even as after-tax income edged lower. That can be confidence. It can be necessity. It can be credit-card gravity. The data do not tell you which one at the grocery aisle.

The first-quarter GDP details are not recessionary, but they are less comforting than the headline from a month ago. BEA said the downward revision reflected weaker investment and consumer spending than previously estimated. Corporate profits rose, but by much less than in the prior quarter. Inflation measures inside GDP remained hot: the quarterly PCE price index increased at a 4.5 percent annual rate, while core PCE rose 4.4 percent.

Markets like to reduce this to the next rate move. That is tidy and insufficient. The more important point is that the economy is absorbing multiple policy shocks at once: energy strain from Middle East instability, tariff uncertainty and refunds, federal spending turbulence, and a consumer still willing to spend but less able to surprise forever.

A cooler growth revision does not buy the Fed much room if prices remain sticky. A hot inflation number does not make rate hikes easy if demand is already losing some heat. The economy has stopped behaving like a dashboard and started behaving like a tug-of-war.

The tariff fight has reached the stage where the revolution is processed through a form.

Reuters reported in late April that the Trump administration expected the first refunds from unlawful emergency tariffs around May 11, after the Supreme Court's February ruling that the president lacked authority to impose sweeping duties under the International Emergency Economic Powers Act. Court documents put the possible refund universe at about $166 billion in duties paid by more than 330,000 importers on roughly 53 million entries.

That is not merely a legal footnote. It is a working-capital event. A business that paid a tariff, passed along part of it, absorbed part of it, repriced inventory, or delayed shipments now has to decide how much of the past belongs on the balance sheet. Refunds do not reverse consumer prices with a magic lever. They move through customs entries, liquidation, Treasury payments, accountants, lenders, and possibly litigation among firms and customers.

The administration has not given up the tariff habit. After the Supreme Court defeat, Trump imposed a new 10 percent global tariff under different authority. A federal trade court ruled against that move in May, though Reuters and AP coverage noted the practical block was narrow, applying to two private importers and the State of Washington rather than wiping the policy off the map.

That is the current shape of American trade policy: legally wounded, tactically persistent, and maddeningly expensive to plan around. Importers cannot simply assume the tariff era is over. They also cannot assume duties they pay today will survive court review tomorrow.

The politics are almost elegant in their confusion. Tariffs are announced as punishment for foreign governments. They are paid first by importers. They are felt by consumers and supply chains. When courts strike them down, the refund flows to the firms that made the entries, not automatically to the families that paid higher shelf prices.

So the public sees a familiar bargain: dramatic executive action, delayed judicial correction, private-sector cash-flow chaos, and only partial repair. A tariff can be a slogan in the morning and a receivable by next quarter. That does not make it strategy. It makes it an invoice with a flag on it.

The United States is not waiting for an AI law to decide how frontier models enter national security work. It is using contracts.

Axios reported last month that the Pentagon reached an agreement with Google allowing use of its model for all lawful purposes, a formulation broader than some rival arrangements and one arriving as Congress remains far from passing military AI guardrails. Google and OpenAI have both said they draw lines against autonomous weapons and mass surveillance. The operational question is what those lines mean once embedded in classified networks, defense workflows, subcontracting chains, and urgent battlefield requests.

The government is also pulling the labs closer before release. Reuters reported this month that Microsoft, Google, and xAI agreed to give the Commerce Department's Center for AI Standards and Innovation early access to new models for national security testing, building on prior arrangements with OpenAI and Anthropic. The stated aim is to evaluate capabilities and security risks before deployment.

That sounds like oversight. It may be part of oversight. But it is also a new intimacy between the state and the model makers. Pre-release access gives government evaluators a window into systems before the public sees them. Defense contracts give labs revenue, status, and a reason to interpret acceptable use in ways that survive procurement pressure.

Congress knows the concern. Axios separately reported that senators sent letters to major AI companies asking about insider access and China-linked safeguards. That is a real issue, but it is not the whole issue. A country can harden models against foreign theft while still failing to decide what domestic democratic control should exist over military AI.

The phrase human in the loop is already aging into ritual language. The question is not whether a human appears somewhere in a process diagram. It is whether that person has time, authority, information, and institutional permission to say no.

For now, the binding text is not a statute debated in public. It is a contract reviewed by lawyers, shaped by agencies, summarized by spokespeople, and discovered by the public after the architecture is already built. That is a poor way to decide where the machine may point.

Cannes did not choose the loudest film in the room. It chose the one that sounds, from a distance, like a family drama and then turns into an argument no dinner table can survive.

Cristian Mungiu's Fjord won the Palme d'Or at the 79th Cannes Film Festival, making the Romanian director one of the rare filmmakers to win the festival's top prize twice. His first Palme came in 2007 for 4 Months, 3 Weeks and 2 Days, a film whose austerity made moral panic feel procedural. Fjord appears to have landed because it found a similarly chilly route into contemporary fracture.

AP described the film as a Norway-set drama about Romanian Evangelicals, played by Sebastian Stan and Renate Reinsve, whose children are taken by child services after they spank them. Mungiu has framed the film as an argument against fundamentalism of more than one kind. The Cannes jury, chaired by Park Chan-wook, rewarded a work about the point at which empathy hardens into state power and family conviction curdles into harm.

That is a very Cannes sentence, yes. It is also the territory culture keeps circling because politics has made ordinary judgment feel radioactive. Who protects children? Who gets to define harm? When does pluralism become a polite word for abandonment? When does intervention become soft coercion in nice shoes?

A useful Palme does not simply identify the year's best film. It tells us what anxieties elite culture is willing to dignify. This year, Cannes dignified polarization without pretending it is only a problem of the crude or the far away. Fjord's cleverness, at least as described from the festival, is that it seems to make northern European liberal order the setting, not the solution.

The award also says something about the exhaustion of easy consensus cinema. A festival can applaud inclusion all day; audiences still know that the hard cases are where slogans go to shed their skin. Mungiu's win suggests Cannes wanted a film that could not be domesticated into a position paper.

Good. The movies should be a little dangerous to people who already know what they think.

The most fashionable lazy take about the 2026 World Cup is that sports are over, attention has migrated, and soccer is now just another legacy spectacle waiting to be outperformed by AI drama and geopolitical panic. This is wrong in the precise way rich people prefer: it mistakes exclusion for decline.

AP reported this week that New York and New Jersey are investigating FIFA's World Cup ticket practices, including variable pricing models and redrawn stadium maps that fans say moved them into worse seats than expected. The tournament begins June 11 and runs through July 19 across the United States, Canada, and Mexico. FIFA's own ticket material advertised group-stage entry points from $60 in an earlier sales phase, while premium final seats were listed in the thousands. The promise was access; the practice looks increasingly like yield management wearing a scarf.

This is not evidence that nobody cares. It is evidence that too many people care, and the event's governors have decided to monetize that care until it squeals. AP reported in January that FIFA said it had received more than 500 million ticket requests. That is not cultural irrelevance. That is demand being converted into resentment.

The pageantry is part of the point. A World Cup in North America was always going to be half tournament, half infrastructure flex: stadiums, airports, security zones, hospitality packages, VIP corridors, sponsor villages, mayoral backslaps, and fans doing math in hotel tabs. The sport is still beautifully simple. The access layer around it has become baroque.

This is why the ticket story matters more than the usual consumer grumbling. FIFA is custodian of a game whose mythology depends on mass belonging. Children in replica shirts are not incidental to the brand; they are the moral alibi. When the path into the stadium feels like an airline fee experiment, the institution is not merely maximizing revenue. It is editing the crowd.

Will people watch? Of course. They will watch in bars, on phones, at work, in languages FIFA's sponsors do not understand. The World Cup remains one of the few remaining global rituals that can interrupt a normal day without asking permission.

But a ritual can be alive and still be degraded. The danger is not that soccer becomes passe. It is that the live version becomes a luxury product while the masses are invited to provide atmosphere from outside the gates. That is not the death of sports. It is gentrification with a kickoff time.

The Iran deal now awaiting Trump's approval has many possible virtues. It may keep ships moving through Hormuz. It may create space for nuclear talks. It may spare people who do not know the names of American lawyers from another round of American ordnance. Take all of that seriously.

Then take this seriously too: Congress is being trained to accept war as a calendar trick.

On May 1, Trump told Congress that hostilities initiated against Iran on Feb. 28 had been terminated, while leaving open future action. The administration's practical position has been that a ceasefire can stop or reset the War Powers clock. Then, this week, the U.S. military said it carried out self-defense strikes in Iran even as negotiations continued. The Senate, for its part, had already blocked a resolution that would have forced the president to end or seek authorization for military action.

This is how constitutional habits die now: not in one grand seizure of authority, but through scheduling ambiguity. The country is not at war, except when it is striking. Hostilities are terminated, except when future hostilities are reserved. Congress has power, except when action is urgent, limited, paused, defensive, separate, deniable, or inconvenient.

The War Powers Act was never a magic shield. Presidents of both parties have treated it as a speed bump with stationery. But the Iran episode exposes a newer degradation. The executive no longer needs to deny Congress a role with any great drama. It can convert military action into a series of episodes and dare legislators to prove continuity.

That matters because continuity is what citizens experience. A sailor deployed near Hormuz does not live in a legal footnote. An Iranian family under the threat of renewed strikes does not inhabit a pause button. A tanker captain deciding whether to move cargo does not care whether Washington calls the last explosion a discrete defensive event.

If Congress wants authority, it has to use authority while use is still possible. It can require authorization for further strikes. It can condition funds. It can demand public reporting. It can stop pretending that a classified briefing is the same as consent.

The pending deal may be good. The method remains rotten. Peace made by one man is brittle. War managed by one man's clock is worse.