The world in March 2026 is characterized by converging crises: democratic erosion in established democracies including the United States; intensifying global conflicts with no clear path to resolution; accelerating climate disruption now firmly past the 1.4°C threshold; AI systems transitioning from experimental tools to embedded infrastructure; and a multilateral order under severe strain.

These are not separate problems — they are interconnected systems feeding back on one another. Here is what the next five years may look like.

The U.S. has paused immigrant visa issuance for nationals of 75+ countries. Net migration turned negative in 2025 for the first time since the 1930s. Interior enforcement has expanded dramatically, with ICE conducting community raids including at schools and courthouses. An estimated 310,000–315,000 removals were executed in 2025, extending into 2026. Humanitarian parole and credible fear protections have been effectively eliminated.

The deterrence-only model will likely produce severe unintended consequences over the next five years. Labor shortages in agriculture, construction, healthcare, and service industries — already visible in 2025–2026 — will become structural. Five years of negative net migration could compound into significant macroeconomic drag.

Globally, the migration pressure won’t disappear — it will reroute. Climate-driven displacement, conflict, and economic collapse in Central America, the Sahel, and South/Southeast Asia will continue to push populations northward and toward Europe. The question by 2031 isn’t whether migration happens, but whether receiving states manage it through orderly systems or through escalating enforcement that generates its own humanitarian crises.

The Committee to Protect Journalists describes U.S. press freedom as “under siege.” Journalist arrests nearly tripled from 15 in 2023 to 49 in 2024, and 2025–2026 has seen further escalation. Federal agents raided a Washington Post reporter’s home and seized her devices. At least nine journalists were assaulted by federal agents in Minneapolis in January 2026. Don Lemon and Georgia Fort were arrested covering protests. Reporters Without Borders documents the first prolonged decline in U.S. press freedom in modern history.

This is one of the most alarming trendlines. The pattern — vilification of media by political leaders, arrest of journalists covering enforcement operations, seizure of devices and source material, removal of reporters from government briefings — follows a well-documented authoritarian playbook. The chilling effects are already measurable: journalists self-censoring, sources refusing to talk, newsrooms downsizing investigative units.

By 2031, expect two press ecosystems to have solidified: a shrinking institutional press operating under increasing legal and financial pressure, and a growing network of independent media, newsletters, and nonprofit journalism outlets that are more agile but less resourced. Local news will continue its collapse — roughly a third of newspapers operating in 2005 have already shuttered, and the trend is accelerating.

The PRESS Act (federal shield law) has failed twice. Without legislative protection, the legal framework for press freedom will continue to erode through precedent — each device seizure, each arrest, each subpoena that goes unchallenged sets a new baseline. AI-generated deepfakes will compound the problem by making it harder for any outlet to establish credibility.

Carnegie Endowment analysts are debating whether the U.S. has already entered “competitive authoritarianism.” The European Parliament’s 2025 human rights report declares democracy “in decline globally.” Foreign Policy describes an “age of defensive democracy” in which democratic backsliding has evolved into a seemingly unstoppable autocratic wave. The U.S. has withdrawn from the UNFCCC, IPCC, and 65 other international organizations.

The November 2026 U.S. midterms will be a critical inflection point — Carnegie calls them a test of “how much resilience U.S. democracy is showing.” If elections are conducted fairly and produce meaningful checks on executive power, the democratic system may stabilize, though with significant institutional damage. If they don’t, the demonstration effects globally will be severe.

The broader picture: the “third wave” of democratization that began in the 1970s has clearly reversed. By 2031, the proportion of the global population living under democratic governance will likely continue declining. What’s new is that established democracies are eroding from within, not collapsing through coups. The mechanisms are incremental: weakened judiciaries, partisan media capture, hollowed-out civil services, normalized norm-breaking.

Counter-pressures exist. South Korea’s democratic resilience (impeaching an attempted coup), Poland’s post-2023 democratic restoration, and mass protests in Serbia and Bangladesh demonstrate that citizens do push back. The question for 2031 is whether these become the pattern or the exception.

AI has transitioned from experimental to infrastructural. Agentic AI systems capable of multi-step autonomous task execution are deployed across industries. Multimodal AI processes text, images, video, voice, and sensor data simultaneously. Deepfakes are now largely indistinguishable from authentic content. Voice cloning enables financial fraud at unprecedented scale. AI World Journal calls 2026 the year “intelligence became infrastructure.”

The next five years will see AI become as ubiquitous as electricity — embedded in every system, often invisibly. AI agents handling complex multi-step workflows will transform white-collar work as fundamentally as industrial automation transformed manufacturing. By 2031, a significant portion of knowledge work will involve human-AI collaboration at minimum, full AI execution in many cases.

The collapse of certainty — AI-generated content becoming indistinguishable from human-created content — will be the defining epistemological crisis of the era. By 2031, no video, audio, or image can be assumed authentic without verification infrastructure that doesn’t yet exist at scale. This cascades into journalism, legal proceedings, elections, and interpersonal trust.

Regulatory fragmentation will intensify. The EU’s AI Act, China’s model-specific regulations, and U.S. industry self-regulation represent divergent paths. By 2031, “regulatory arbitrage” becomes standard practice. The concentration of AI capability in 5–8 major companies and two countries (U.S. and China) will deepen. Whether AI development serves public interest or shareholder interest becomes the defining technology policy debate.

AI-powered surveillance has become foundational to both state security and commercial operations. The Trump administration has deployed facial recognition, AI social media scraping, and robotic patrol systems for immigration enforcement. Autonomous AI agents are reshaping video surveillance from reactive monitoring to predictive risk identification. Security systems now detect “pre-incident” patterns.

Surveillance in 2031 will be qualitatively different. The convergence of ubiquitous high-resolution cameras, AI behavioral analysis, IoT sensor networks, biometric identification, and predictive analytics will create what researchers call “ambient intelligence”: environments that continuously observe, analyze, and respond to human behavior.

In immigration enforcement specifically, expect AI-integrated surveillance to extend well beyond the border — into workplaces, transit systems, and digital communications. The EFF and UAW’s lawsuit challenging government social media surveillance signals the legal battles ahead, but the technology is advancing faster than legal frameworks.

The last three years have all exceeded 1.4°C above preindustrial levels. 2024 was the hottest year in 175 years of record-keeping. Atmospheric CO₂ reached 152% of 1750 levels. The 2023–2025 mean hit +1.5°C — the Paris Agreement’s aspirational limit. The U.S. has withdrawn from the Paris Agreement (again), the UNFCCC, the IPCC, and the Green Climate Fund. U.S. emissions reversed their downward trend and increased in 2025.

By 2031, the 1.5°C threshold will have been definitively breached as a multi-year average. The WMO projects each year from 2025–2029 at 1.2–1.9°C above preindustrial. James Hansen projects temperatures could reach +1.7°C by 2027, with continued rise thereafter.

There is a deep paradox. Clean energy deployment is accelerating — over 90% of new U.S. power capacity in 2025 was renewable, driven by economics not policy. China is positioning itself as the clean energy superpower. EVs are already cheaper to buy than conventional cars in China. Wind and solar provided 30% of EU electricity in 2025, surpassing fossil fuels for the first time. But none of this is happening fast enough. Scientists estimate emissions need to fall ~50% by 2030. That is not going to happen.

By 2031, adaptation will dominate the climate discourse at least as much as mitigation. Cities — particularly in the Global South — will face existential heat, flooding, and water stress. Climate-driven migration will intensify. China’s clean technology dominance will reshape the global economic balance of power while the U.S. doubles down on fossil fuels.

The U.S. has withdrawn from or defunded numerous multilateral institutions. The Security Council remains paralyzed by great-power vetoes. COP30 in Belém failed to endorse a fossil fuel phase-out. The $1.3 trillion climate finance goal has no clear implementation pathway. The ICRC warns of converging trends pushing toward “deeper instability and human suffering.”

Foreign Policy‘s analysis is instructive: the UN will likely remain central because it validates sovereignty — the attribute all states prize most — but its authority and relevance will decline further. None of the P5 wants to see the UN displaced, but none is willing to make it effective either.

By 2031, expect: institutional persistence without institutional effectiveness; proliferating regional arrangements (BRICS+, bilateral pacts, trade blocs) operating alongside but increasingly outside UN frameworks; and a growing humanitarian financing gap. International humanitarian law will persist in theory while being so consistently violated in practice that its deterrent function is effectively gone.

Multiple high-intensity wars continue: Russia-Ukraine (year 4), Israel-Palestine/Gaza (60,000+ Palestinian deaths, the strip largely destroyed), Sudan’s civil war (20,000+ killed in 2024–25, 12 million displaced), Myanmar’s civil war, and escalating violence in Haiti, Mexico, Ecuador. A brief India-Pakistan conflict occurred between nuclear powers in May 2025. The ICRC is working in more than 100 armed conflicts. The Council on Foreign Relations identifies growing domestic political violence in the United States as a Tier I concern.

The Russia-Ukraine war will likely continue in some form — active combat, frozen conflict, or negotiated settlement. Europe’s security architecture has been permanently altered. The Middle East will remain volatile, with the underlying questions of Palestinian statehood, Israeli security, and Iranian nuclear ambitions unresolved. Sudan is the world’s most underreported catastrophe and could see death tolls rivaling the worst humanitarian disasters of the century.

The broader trend: war is becoming more common, more civilian-targeting, and more resistant to diplomacy. Defense spending is rising globally while peace-building infrastructure is being dismantled. By 2031, the number of active armed conflicts will likely remain at or above current historically elevated levels.

The most honest assessment is that we are living through a period of systemic transition where the old order is dying and the new one has not yet been born. Several of the trends above are mutually reinforcing in troubling ways:

But counter-trends also exist: clean energy economics are increasingly favorable regardless of policy; citizen movements continue to push back against democratic erosion; independent media networks are growing even as institutional press shrinks; international humanitarian law, though weakened, still provides frameworks for accountability.

By 2031, the world will not have collapsed — but it will be significantly more unequal, more surveilled, more ecologically stressed, and less democratically governed than it was in 2020. The crucial variable, as always, is what organized people choose to do in the face of these forces.