Global Foresight 2036 is the fifth edition of the Atlantic Council’s annual strategic‑foresight series. It combines a worldwide survey of ≈ 450 geostrategists, a set of six “snow‑leopard” under‑the‑radar trends, and expert video commentary on artificial intelligence. The report’s aim is to surface the drivers of change that could shape geopolitics, technology, economics, and governance over the next decade.
Purpose & Methodology
- Goal – Produce provocative, non‑partisan insights about the world in 2036 and help policymakers test assumptions.
- Survey – Fourth annual questionnaire (Nov‑Dec 2025) of 447 experts from 72 countries covering geopolitics, economy, governance, science & tech, climate, and security.
- Respondent profile – ≈ 50 % U.S., > 20 % Europe, ≈ 20 % Global South; 75 % male; ≈ 70 % over 50 years old; mixed sectors (private, NGOs, government, academia, multilateral).
Survey Findings – Ten Biggest Themes
| Key Finding | Representative data | Expert nuance |
|---|---|---|
| China’s economic rise & Taiwan risk | 70 % think China will try to seize Taiwan by force (2026); 58 % expect China to be the world’s top economic power by 2036. | Dexter Tiff Roberts warns China’s export‑driven model may stall; Markus Garlaskas notes military power alone won’t deter Beijing. |
| NATO’s uncertain future | 44 % believe NATO will not exist in its current form in 2036; 56 % think it will survive. | Ian Brzezinski stresses U.S. leadership as essential; 39 % doubt the U.S. will retain a central NATO role. |
| Russia’s frozen conflict in Ukraine | 52 % foresee a frozen conflict; only 34 % expect a Russian‑favorable outcome. | John Herbst argues sustained U.S. pressure remains key to a durable peace. |
| AI’s dual impact | 56 % see AI as overall positive, but 15 % view AI disruption as the biggest threat to prosperity (2026). | Tess Deblanc‑Knowles cautions that AGI may never materialise under current trajectories. |
| Nuclear proliferation | 85 % expect new nuclear actors; Iran (66 %) and Saudi Arabia (53 %) top the list. | Amy Woolf highlights the paradox of growing arsenals yet low expectations of use. |
| Europe’s “strategic autonomy” | 57 % agree Europe will achieve strategic autonomy by 2036 (up from 31 % in 2024). | Jörn Fleck describes simultaneous abandonment and entrapment by the U.S. |
| Climate‑driven water wars | Only 17 % now cite climate change as the biggest threat (down from 30 % in 2024); 64 % expect conflicts over fresh water. | Kathleen Euler notes that climate “remoteness” hampers decisive action. |
| Decaying international institutions | 71 % foresee the UN becoming less influential; similar pessimism for WTO, IMF, World Bank. | Matthew Kroenig links institutional decay to a broader democratic recession. |
| Dollar’s dominance challenged | 34 % see cryptocurrency as the biggest challenger; 21 % point to the renminbi. | Alisha Chhangani notes that crypto’s volatility and regulatory uncertainty keep the dollar’s base strong. |
| Global South perspective diverges | 76 % of Global South respondents think China will surpass the U.S. economically (vs 54 % of others); 22 % expect the U.S. to break up (vs 10 % elsewhere). | James Mazzarella attributes this to perceived U.S. isolation and historic fragility of liberal institutions. |
Key quantitative take‑aways
- World outlook: 63 % expect the world to be worse off in 2036; only 37 % think it will be better.
- Great‑power war: 41 % see a new world war possible; the most‑cited flashpoint is Taiwan/East China Sea (43 %).
- AGI by 2036: 58 % believe artificial general intelligence will be achieved.
- AI in lethal systems: 76 % expect autonomous weapons to be deployed without a human‑in‑the‑loop.
Snow Leopard Trends (2022‑2026)
The Atlantic Council’s “snow leopards” spotlight low‑visibility but high‑impact phenomena. 2026’s six are:
| Snow leopard | Why it matters |
|---|---|
| Tech firms entering conflicts | Private‑sector cyber capabilities are now acting as de‑facto combatants, raising legal and strategic risks. |
| Brain‑circulation migration | Skilled diaspora returning home fuels entrepreneurship and regional innovation loops. |
| Kelp‑forest climate solutions | Seaweed removes CO₂ 10‑times more efficiently than land forests; offers carbon‑negative, bio‑economic pathways. |
| Neurotechnology mind‑reading | Emerging semantic decoders could translate brain activity into text, with medical benefits but profound privacy concerns. |
| AI‑generated cultural erasure | Western‑centric data sets risk marginalising non‑English cultures, potentially rewriting collective memory. |
| Sovereign AI regimes | Nations will push for AI models aligned to domestic law, security, and values—challenging global AI governance. |
AI Expert Opinions
Three Atlantic Council technology experts discussed AI’s trajectory:
| Expert | Core message |
|---|---|
| Tess Deblanc‑Knowles | Current LLMs excel at next‑token prediction; they cannot generate truly novel knowledge—AGI may be farther away than hype suggests. |
| Emerson Brooking | AI is probabilistic, not predictive; the real power will emerge when AI augments human forecasters rather than replaces them. |
| Trey Herr | Trust gaps (≈ 60 % of U.S. adults distrust AI outputs) could trigger a market cool‑down if high‑profile accidents occur. |
| Graham Brooke | AI will continue to transform routine domains (weather, traffic) but will need “world models” to impact broader societal tasks. |
Conclusions & Recommendations
- Strategic vigilance – Policymakers should monitor the Taiwan flashpoint, NATO’s internal cohesion, and the accelerating nuclear ambitions of Iran and Saudi Arabia.
- Invest in resilience – Strengthen allied intelligence, intelligence‑sharing, and rapid‑response doctrines to deter both conventional and cyber‑enabled aggression.
- Guide AI governance – Prioritise transparency, trust‑building, and international frameworks that prevent sovereign AI fragmentation while safeguarding democratic values.
- Leverage “snow leopards” – Support public‑private cybersecurity coordination, incentivise diaspora return programmes, fund kelp‑restoration research, and establish ethical standards for neuro‑tech and AI‑generated culture.
Prepared by Peter Engelke, Senior Fellow for Foresight, Scowcroft Center for Strategy and Security – Atlantic Council.