# 5 Scenarios for How the World Will Govern AI by 2030

By 2030, global artificial intelligence governance will bypass a single, unified rulebook in favor of fractured, competing regional models. While the European Union champions strict risk-based compliance and the United States pivots toward market-driven deregulation, the Global South is rapidly building sovereign frameworks to avoid digital dependency. Ultimately, whether AI's future resembles a collaborative diplomatic utopia or a fragmented regulatory arms race depends entirely on how competing nations reconcile algorithmic safety with the economic pressure to innovate.

## The Splintering of the Global AI Consensus

For years, a quiet assumption ran through the global technology industry: eventually, governments would sort out AI regulation, and a single, universal playbook would emerge [cite: 1]. Observers anticipated a "Brussels Effect" for artificial intelligence—a scenario where the European Union's regulatory framework would become the de facto global standard, much like the General Data Protection Regulation (GDPR) did for data privacy [cite: 2, 3]. 

As the world moves through 2026, that assumption has entirely collapsed. The regulatory landscape is not converging; it is fracturing into a mosaic of sovereign AI bubbles [cite: 1, 4]. AI's impact remains undeniably global, but its rules are intensely local, driven by competing national interests, geopolitical rivalries, and deeply embedded cultural values [cite: 4]. 

This fragmentation is no longer a theoretical risk; it is a present reality defined by regional lockouts and compliance gridlocks. The most glaring example is the geofencing of consumer AI tools. When Apple rolled out "Apple Intelligence," the company explicitly blocked the suite of advanced AI features from devices in two of the world's largest markets: the European Union and mainland China, which together represent nearly two billion people [cite: 5, 6]. 

In the EU, the lockout was driven by a standoff over the Digital Markets Act (DMA). European regulators classify Apple as a "gatekeeper," requiring strict interoperability with third-party systems—a demand Apple argued would compromise its on-device privacy architecture [cite: 7, 8]. Meanwhile, in China, the blockade stems from rigid regulations requiring state approval for large language models (LLMs) and mandatory data localization, combined with national bans on foreign AI models like ChatGPT [cite: 7, 8]. 

These distinct regulatory environments highlight a critical reality for enterprise technology: businesses are no longer building one AI product for the world. They are being forced to engineer parallel compliance architectures, and in some cases, entirely different product pipelines, to navigate the widening gap between state-led control, aggressive risk mitigation, and deregulated innovation [cite: 1, 9]. 

### Exposing the Myths That Sabotage Governance

The fracture in global policy is exacerbated by persistent misconceptions regarding how AI actually operates and where its risks lie. Regulators and corporate boards are increasingly realizing that the "innovation privilege" that shielded early AI development has expired [cite: 10]. For years, speed mattered most, experimentation was universally encouraged, and failures were written off as the cost of technological progress [cite: 10]. 

By 2026, regulators have dismantled the justifications that once protected tech firms. The most prominent casualty is the "black box" defense. Historically, companies claimed that advanced AI systems were simply too complex to fully understand or explain [cite: 10, 11]. Today, international regulatory guidance explicitly rejects this argument. If an organization claims its system is too opaque to be governed, regulators deem it too opaque to be deployed in high-impact contexts [cite: 10]. Explainability and human-in-the-loop oversight are no longer aspirational goals; they are strict legal mandates in heavily regulated sectors like financial services and healthcare [cite: 10, 11].

Furthermore, governance frameworks are moving away from the myth that AI is purely a mathematical or coding issue. Framing AI as purely technical allows organizations to externalize responsibility. In reality, AI systems are socio-technical systems [cite: 12]. They are shaped by human choices regarding data selection, optimization targets, deployment contexts, and acceptable error rates [cite: 12]. When an AI system is used for employment screening or administrative triage, the risk rarely lies in the algorithm alone; it lies in the institutional decisions surrounding how the system is relied upon [cite: 12].

Regulators are also actively countering the assumption that bigger datasets automatically produce better, safer AI. In regulated contexts, massive but poorly curated datasets can amplify noise, bias, and spurious correlations [cite: 12]. Consequently, high-quality, representative datasets are being legally prioritized over sheer scale, forcing companies to prove their data provenance [cite: 12]. Fiduciary risk now explicitly covers AI; if an AI model hallucinates or demonstrates algorithmic bias leading to consumer harm, the organization deploying the tool is fully liable, regardless of whether they built the model or purchased it from a third party [cite: 10, 11].

## Three Competing Gravity Wells of Governance

The current geopolitical landscape of AI governance is dominated by three major regulatory philosophies. Each framework exerts its own gravitational pull on global markets, forcing multinational corporations to choose which rules they will follow and which markets they will abandon [cite: 4].

### The European Union and the Limits of the Brussels Effect
The European Union enacted the world's first comprehensive legal framework for AI, the EU AI Act, which entered into force in August 2024 and phases in its most consequential enforcement provisions through 2026 [cite: 3, 13]. The legislation is built on a tiered, risk-based approach that heavily penalizes non-compliance, with fines reaching up to 35 million Euros or seven percent of global turnover for prohibited practices [cite: 9]. 

The EU categorizes AI systems based on their potential to cause harm. "Unacceptable risk" systems—such as government social scoring, manipulative subliminal surveillance, and real-time facial recognition in public spaces by law enforcement—are outright banned [cite: 3, 13]. "High-risk" applications, spanning employment screening, biometric identification, and critical infrastructure management, face stringent conformity assessments, continuous monitoring, and fundamental rights impact assessments prior to market placement [cite: 9, 13]. 

By August 2025, obligations for general-purpose AI (GPAI) models commenced, accompanied by a comprehensive Code of Practice [cite: 13]. However, the EU's attempt to lead the world by establishing the "gold standard" has encountered significant friction. The EU envisions its regulations as a trust-building exercise that will foster sustainable, human-centric innovation [cite: 3, 14]. Critics, however, argue that the compliance burden creates a "Brussels Side-Effect" [cite: 15]. Because the AI Act functions more like a traditional product safety regulation rather than a dynamic technological framework, it risks stifling domestic AI development and incentivizing multinational companies to launch products elsewhere [cite: 14, 15]. This tension peaked when major American tech executives publicly stated their refusal to sign the GPAI Code of Practice, citing legal uncertainties and regulatory overreach, thereby highlighting the limits of the EU's persuasive power [cite: 13].

### The United States and Market-Driven Deregulation
In stark contrast to the European Union, the United States has adopted a highly fragmented, market-driven approach that prioritizes rapid innovation and geopolitical competitiveness [cite: 4, 9]. Historically, the U.S. regulatory environment relied on voluntary commitments from major tech companies, combined with targeted sector-specific guidelines [cite: 4, 16].

The U.S. trajectory shifted decisively in the latter half of 2025. Following the unveiling of a comprehensive strategy titled "Winning the Race: America's AI Action Plan," the federal government explicitly focused on accelerating AI infrastructure, securing U.S. leadership, and countering foreign adversaries [cite: 17]. This culminated in sweeping executive actions, such as Executive Order 14365, which aimed to decrease barriers to innovation and establish federal preemption over restrictive state-level laws [cite: 4, 9]. 

The prevailing American philosophy treats artificial intelligence primarily as a strategic economic and military asset that must be shielded from heavy-handed government interference [cite: 4]. While certain states, such as Colorado and Texas, have attempted to pass their own AI governance acts focusing on algorithmic discrimination and documentation, federal policy actively discourages compliance with state laws that might slow down the rapid deployment of frontier models [cite: 9, 18]. This dynamic creates a chaotic compliance environment for multinationals operating within the U.S., as they must navigate conflicting state and federal priorities without a unified national framework [cite: 9].

### China and Strategic Sovereign Control
China views artificial intelligence as a critical pillar of both national power and internal social stability. Its approach is heavily state-led, focusing on stringent content censorship, algorithmic registration, and absolute alignment with state interests [cite: 4]. 

At the 2025 World Artificial Intelligence Conference (WAIC) in Shanghai, Beijing showcased its "Global AI Governance Action Plan," attempting to position itself as a global leader in establishing international AI norms [cite: 17]. China's strategy outwardly emphasizes global solidarity, open-source communities, and green AI technologies. Simultaneously, it enforces rigid domestic rules that mandate local data storage, restrict foreign AI systems, and require that all generative AI outputs reflect socialist core values [cite: 7, 17]. 

China explicitly ties its AI governance to its broader foreign policy objectives. The nation offers capacity-building assistance, digital infrastructure development, and technology transfers to developing nations in the Global South [cite: 17]. This initiative is designed to bridge the AI divide while establishing a diverse, China-friendly technological ecosystem that relies on Chinese hardware and governance models rather than Western alternatives [cite: 17].

| Feature | European Union | United States | China |
| :--- | :--- | :--- | :--- |
| **Core Regulatory Philosophy** | Rights-respecting, risk-based regulation emphasizing safety. | Market-driven, deregulated innovation prioritizing speed. | State-controlled, strategic national asset emphasizing stability. |
| **Primary Mechanism** | The EU AI Act (Tiered risk obligations and heavy fines). | Voluntary frameworks, fragmented state laws, Executive Orders. | Centralized algorithmic registry, strict data localization. |
| **Stance on Innovation** | Trust and safety are absolute prerequisites for sustainable innovation. | Minimal government interference; prioritize market dominance. | AI must be tightly aligned with state stability and economic planning. |
| **Global Influence Goal** | The "Brussels Effect" (Exporting regulatory standards globally). | Exporting frontier technology and maintaining geopolitical supremacy. | Shaping alternative global norms; capacity building in developing nations. |

## The Rise of the Global South in AI Policy

Perhaps the most significant macro-trend in global AI governance between 2024 and 2026 has been the aggressive emergence of the Global South as a unified regulatory force. Historically, low- and middle-income countries have been marginalized in international technology governance. They have operated as "rule-takers," supplying training data, essential minerals, and deployment sites without reaping commensurate economic benefits or securing adequate ethical safeguards [cite: 16]. 

This structural post-colonial dependency is rapidly changing. Burned by inequitable digital structures in the past, countries across the Global South—urged by institutions like the World Bank—are preemptively establishing sovereign frameworks. These initiatives are designed to ensure their socio-economic priorities, technological sovereignty, and cultural values are embedded in the global AI race [cite: 16, 17, 19].

### India’s Techno-Legal Framework and "AI for All"
India has assertively positioned itself as the leader of the Global South in technology governance, pursuing a "techno-legal" approach that balances rapid innovation with strict accountability [cite: 20]. India's strategy involves a total transition away from the outdated Information Technology Act of 2000 toward the comprehensive Digital India Act of 2025 [cite: 21, 22]. 

The Digital India Act represents a massive overhaul of Indian cyberspace law. It establishes strong algorithmic accountability, specific regulations for synthetic media and deepfakes, and robust user safety mechanisms, all while heavily promoting the expansion of Digital Public Infrastructure (DPI) [cite: 21, 22, 23]. India's Ministry of Electronics and Information Technology (MeitY) has also published comprehensive AI Governance Guidelines, resting on principles of human-centric design, equity, and transparency [cite: 20]. Furthermore, India has debated an Artificial Intelligence (Ethics and Accountability) Bill, which proposes mandatory ethical reviews for high-risk systems and statutory bias audits, signaling a willingness to enforce strict penalties for corporate negligence [cite: 20].

### The 2026 New Delhi Declaration and Accountability Compact
India's leadership reached a critical inflection point on the global stage in early 2026. The international AI summit circuit, which began with a focus on existential risks in the Global North, fundamentally shifted its geography and its priorities. 

The first AI Safety Summit at Bletchley Park in 2023 generated a shared international vocabulary around frontier-AI risks [cite: 19, 24]. The 2024 Seoul Summit expanded this by committing to a network of AI safety institutes [cite: 19, 25]. The 2025 Paris AI Action Summit marked a shift away from pure safety toward private investment and environmental sustainability [cite: 19, 26, 27]. However, it was the February 2026 AI Impact Summit in New Delhi that truly reframed the global agenda [cite: 19, 28].

[image delta #1, 0 bytes]

 

The New Delhi summit resulted in the "New Delhi Declaration on AI Impact," a landmark agreement endorsed by 88 nations and international organizations—including major powers like the U.S., China, the UK, and the EU [cite: 29, 30, 31]. Based on the principle of *Sarvajan Hitaya, Sarvajan Sukhaya* (Welfare for All, Happiness for All), the declaration shifted the focus from existential fear to practical, equitable distribution of AI benefits [cite: 29, 32]. 

The declaration established seven pillars for global cooperation. It prioritized the democratization of AI resources, ensuring affordable access to foundational compute power for developing nations [cite: 29, 33]. It emphasized AI for economic growth, secure and trusted AI benchmarks, human capital development through vocational training, and resilient, energy-efficient AI systems [cite: 29, 30, 33]. To support these pillars, the summit launched the Global AI Impact Commons (to scale successful AI applications) and a Charter for the Democratic Diffusion of AI [cite: 32, 33].



Months later, on May 14, 2026, India deepened its regulatory footprint by hosting the inaugural International AI Accountability Forum [cite: 34]. This gathering resulted in the adoption of the *New Delhi Compact on Artificial Intelligence Accountability*, an operational framework designed to tackle the impending risks of autonomous Agentic AI [cite: 34, 35]. A core component of this compact is the Universal Declaration of AI Accountability Rights, a foundational document tracking alongside human rights frameworks to establish clear liability chains and compliance requirements for AI-driven harms, setting a target for an International AI Accountability Convention by 2030 [cite: 34, 35].

### The African Union’s Continental AI Strategy
Simultaneously, the African continent has moved to establish its own unified digital perimeter. In July 2024, the African Union (AU) Executive Council formally endorsed the Continental Artificial Intelligence Strategy [cite: 36, 37]. Guided by the aspirations of Agenda 2063 and the Malabo Convention, the strategy explicitly rejects a one-size-fits-all Western regulatory model, opting instead for an "Africa-centric" approach [cite: 36, 38]. 

The strategy focuses on five core themes: maximizing the benefits of AI for economic growth, building continental capabilities and digital infrastructure, minimizing risks, stimulating public and private investment, and fostering international cooperation [cite: 36, 39]. The AU prioritizes specific sectors for immediate AI integration, including agriculture, healthcare, education, and climate change adaptation [cite: 39]. 

A central pillar of the AU's strategy is data sovereignty [cite: 38]. By early 2025, AU leaders officially recognized data sovereignty as a fundamental principle, ensuring that AI development on the continent relies on local data processed within local borders [cite: 38]. The strategy's initial implementation phase (2025-2026) focuses on mobilizing resources, establishing AI advisory boards, and assisting member states in drafting national policies to prevent regulatory fragmentation across its 55 member states [cite: 40].

### The UAE’s Systemic Governance Transformation
The United Arab Emirates represents a uniquely aggressive and highly capitalized model of sovereign AI development. Driven by the UAE National Strategy for Artificial Intelligence 2031, the nation is not merely regulating AI as an external corporate tool; it is embedding AI into the architectural DNA of the state itself [cite: 41, 42]. Abu Dhabi has committed significant capital, deploying billions to become the world's first fully AI-native government across all digital services by 2027 [cite: 43].

The UAE’s approach is exemplified by G42, an Abu Dhabi-based AI powerhouse [cite: 44]. G42’s massive partnerships, including significant infrastructure deals with Microsoft to build sovereign data centers, signal a shift from theoretical governance to "executable governance" [cite: 42, 44]. Rather than debating ethics in white papers or passing restrictive compliance laws, the UAE is building scalable, sovereign AI infrastructure that fuses national interests with data autonomy [cite: 42]. This model proves that a state can implement highly advanced AI systems without compromising on sovereign control, allowing the UAE to actively shape global standards by exporting infrastructure rather than ideology [cite: 42].

| Region | Primary AI Governance Framework | Strategic Focus | Approach to Sovereignty & Innovation |
| :--- | :--- | :--- | :--- |
| **India** | Digital India Act 2025 & New Delhi Compact | "AI for All," algorithmic accountability, and DPI expansion. | "Techno-legal" balance; demands equitable access and strict liability for autonomous agents. |
| **African Union** | Continental AI Strategy (2024) | Data sovereignty, infrastructure building, and cultural preservation. | Rejects Western dependency; prioritizes local data centers and sector-specific AI (agriculture, health). |
| **UAE** | UAE National AI Strategy 2031 | Becoming an AI-native government; executable governance. | Fuses state infrastructure with AI; relies on heavy capital investment and strategic corporate alliances (e.g., G42). |

## Forecasting the Future: 5 AI Governance Scenarios for 2030

Forecasting the exact state of AI governance by 2030 requires analyzing massive critical uncertainties. Chief among these are the speed of AI capability growth, the concentration of corporate market power, and the geopolitical willingness to cooperate [cite: 45]. 

By synthesizing comprehensive foresight reports from the UK Government Office for Science (GO-Science) and the Centre for Future Generations (CFG), five distinct, highly plausible scenarios emerge for the year 2030 [cite: 45, 46, 47]. These scenarios are mapped along two primary variables: the concentration of power (ranging from highly decentralized open-source ecosystems to highly centralized monopolies) and the speed of capability growth (ranging from a stagnant plateau to rapid, explosive advancement) [cite: 46]. These scenarios are not mutually exclusive predictions, but rather stress-test models illustrating how interacting uncertainties might shape global policy [cite: 45].

| Scenario | Concentration of Power | Speed of Capability Growth | Core Characteristic |
| :--- | :--- | :--- | :--- |
| **1. The Splinternet of AI** | Highly Centralized (State-Controlled) | Rapid / Explosive | A geopolitical arms race driven by mistrust and digital protectionism. |
| **2. Big Tech Oligopoly** | Centralized (Corporate Monopolies) | Rapid / Explosive | Private tech giants capture global governance; labor markets are devastated. |
| **3. The AI "Wild West"** | Highly Decentralized (Open Source) | Rapid / Explosive | Democratized innovation leads to unmanageable cybercrime and digital chaos. |
| **4. Cooperative Diplomacy** | Centralized (International Coalitions) | Moderate | A UN-backed licensing regime ensures safety but entrenches incumbent firms. |
| **5. The AI Winter** | Decentralized / Moderate | Gradual / Plateau | AI hits a technical ceiling; regulatory panic subsides into standard software law. |

### Scenario 1: The Splinternet of AI (Geopolitical Arms Race)
In this scenario, advanced AI is viewed entirely through the lens of national security and economic supremacy. The United States and China engage in a fierce, zero-sum competition for AI dominance, heavily guarding semiconductor supply chains and physical compute clusters [cite: 46]. 

Because there is no reliable way to verify an adversary's AI capabilities or safety protocols, mistrust paralyzes international cooperation [cite: 46]. The world fragments into rigid geopolitical blocs. Countries enact aggressive data localization laws, digital firewalls, and strict import bans, severely restricting the cross-border flow of AI models [cite: 1, 4]. Multinational corporations are forced to build entirely separate, redundant AI systems for North America, Europe, and Asia to comply with incompatible laws. In the worst iteration of this scenario, the race ends in a fragile equilibrium punctuated by tense standoffs, or escalating military conflicts driven by autonomous systems and disputes over critical hardware infrastructure [cite: 46].

### Scenario 2: Big Tech Oligopoly (The Agent Economy)
Here, the development of Frontier AI is incredibly fast, but astronomically capital-intensive. The computational costs required to train the next generation of models are so high that only a handful of massive tech conglomerates—primarily based in the U.S. and China—can afford to build and deploy advanced AI agents [cite: 46]. 

These corporations effectively capture global governance. While national governments attempt to pass regulations, the sheer speed of technological change renders state-led laws obsolete before they are even enforced [cite: 4]. In this "Agent Economy," tireless digital workers power explosive economic growth and unprecedented corporate profits, but labor markets are devastated, leading to deep income inequality [cite: 45, 46]. Governance is not dictated by sovereign laws, but by the Terms of Service of private companies. This concentration of power leads to "Silicon Blackmail," where tech giants become so powerful they can effectively challenge or dictate terms to sovereign nations under the threat of withdrawing vital digital infrastructure [cite: 46].

### Scenario 3: The AI "Wild West" (Decentralized Mayhem)
In the mid-to-late 2020s, an algorithmic breakthrough drastically reduces the compute power required to train highly capable models [cite: 45]. Suddenly, open-source AI agents proliferate globally, available to anyone with a standard commercial computer.

This democratization unleashes massive grassroots innovation, destroying the monopolies of Big Tech and distributing AI's benefits widely [cite: 45]. However, it also completely disables regulatory enforcement. Bad actors, authoritarian states, and individual cybercriminals use these decentralized tools to execute large-scale financial fraud, synthesize bioweapons, clone identities, and launch autonomous cyber-attacks [cite: 45, 46]. Governments scramble to react, but the proliferation is impossible to contain. The global economy suffers heavy losses due to widespread digital distrust, and society fractures as citizens can no longer agree on baseline objective reality amidst a relentless flood of synthetic media and deepfakes [cite: 45].

### Scenario 4: Cooperative Diplomacy (Licensed Utopia)
Driven by a highly visible, near-miss catastrophic event involving an advanced AI system, the international community wakes up to the shared existential risks of the technology [cite: 46]. Panicked governments prioritize safety and human control over economic competition.

The United Nations and global coalitions successfully broker a binding international framework. This regime relies on a globally verified licensing system for high-compute training runs, mandatory cross-border auditing, and strict safety benchmarks [cite: 46]. While this scenario represents a diplomatic utopia where AI is made safe and its benefits are shared, it carries a significant economic trade-off. The rigorous licensing regime creates a massive barrier to entry, allowing a few licensed incumbent firms to corner the global market, thereby stifling grassroots open-source development and centralizing technological power within a slow-moving bureaucracy [cite: 46].

### Scenario 5: The AI Winter (Governance Plateau)
In this scenario, the highly anticipated leap to Artificial General Intelligence (AGI) simply does not happen. Generative models hit a "data wall," running out of high-quality human-generated text to train on, and synthetic data proves ineffective at pushing capabilities further [cite: 45, 46]. AI systems struggle with complex, multi-step tasks, remaining useful but narrow tools for specific administrative or coding automation [cite: 45, 46].

As productivity improvements remain only marginal, investors grow disillusioned. Capital flees the AI sector for other emerging technologies, such as fusion energy or quantum computing [cite: 45]. Consequently, the frantic political push for unprecedented global governance cools. The regulatory landscape normalizes, treating AI like any other enterprise software. Regulatory bodies revert to existing consumer protection, anti-discrimination, and data privacy laws to manage the moderate, mundane risks of AI, abandoning the costly pursuit of novel global treaties [cite: 45].

## Lessons from the Past: Analogies for Global Governance

As policymakers struggle to define the architecture of international AI governance, they frequently look to history for institutional blueprints to manage the risks and opportunities of transformative technology. The UN High-level Advisory Body on AI has explicitly emphasized the need for a globally distributed architecture, prompting intense debates over which historical model best fits the unique characteristics of artificial intelligence [cite: 48, 49]. 

### The IAEA Model: The Nuclear Analogy
Many advocates, including leading technology CEOs and prominent academics, suggest establishing an international agency modeled after the International Atomic Energy Agency (IAEA) [cite: 50, 51]. This model would involve a supranational body tasked with monitoring highly capable, potentially dangerous AI systems through strict licensing, non-proliferation treaties, and physical inspections of advanced data centers [cite: 50]. 

However, critics argue that the nuclear analogy is fundamentally flawed. Nuclear technology relies on highly trackable, physical raw materials like enriched uranium, which require massive, detectable industrial facilities to process [cite: 50, 52]. AI, conversely, is built on diffuse, rapidly evolving software and data. Once an algorithm is developed, it can be duplicated and distributed globally in seconds via the internet, making physical containment and traditional non-proliferation strategies nearly impossible to enforce against non-state actors [cite: 50, 52].

### The ICAO and IMO Models: Jurisdictional Certification
Given the difficulty of controlling software, others suggest adapting the frameworks used for international civil aviation (ICAO) or maritime shipping (IMO) [cite: 53, 54]. These models do not attempt to inspect every single airplane or ship globally; instead, they operate on a system of jurisdictional certification. 

Under this analogy, an "International AI Organization" (IAIO) would not audit individual AI models or private tech companies. Rather, it would audit *nations* to ensure their domestic laws and regulatory bodies meet internationally agreed-upon minimum safety standards [cite: 54]. Member states would then give force to these standards by adopting domestic regulations that ban the import of AI products—or restrict access to computing hardware—from non-certified jurisdictions [cite: 54]. This model respects national sovereignty while utilizing global trade mechanics to enforce compliance.

### The CERN Model: Collaborative Public Infrastructure
For developing collective capabilities rather than pure regulation, the CERN (European Organization for Nuclear Research) model is highly attractive. Rather than focusing on restricting AI, an international, publicly funded mega-laboratory for AI could pool global scientific talent and computational resources [cite: 52, 53]. This public institution would build open, safe, and transparent foundational models, counterbalancing the secretive dominance of private Big Tech laboratories and ensuring equitable access to cutting-edge research for developing nations [cite: 52].

The consensus among experts in 2026 is that AI governance cannot rely on a single historical analogy. A successful framework will likely require a polycentric hybrid: applying IAEA-style strict oversight for the few massive compute clusters capable of training frontier models, ICAO-style jurisdictional standards for cross-border software trade, and CERN-style collaboration to ensure equitable access to innovation for the Global South [cite: 52].

## The Role of the United Nations and Multilateralism

The United Nations has positioned itself as the primary vehicle for coordinating these disparate global efforts, aiming to prevent the total fragmentation of AI governance. In late 2024, the UN Secretary-General’s High-level Advisory Body on Artificial Intelligence released its final report, "Governing AI for Humanity" [cite: 48, 55]. The report starkly highlighted a "global governance deficit," noting that current initiatives exclude entire regions and risk creating incompatible regimes [cite: 56].

The UN report outlined a blueprint for a globally inclusive and distributed architecture, proposing seven key recommendations. These included establishing an International Scientific Panel on AI to create a baseline of objective knowledge, initiating a policy dialogue on governance, creating an AI standards exchange, and developing a global AI capacity development network [cite: 49, 55]. Crucially, to address the structural exclusion of the Global South, the UN recommended establishing a Global Fund for AI and a Global AI Data Framework to facilitate equitable access to resources [cite: 49, 55]. 

However, the UN’s multilateral approach faces severe headwinds. Critics argue that while the UN possesses unmatched convening power, it lacks the agility to keep pace with rapid technological advancements and the enforcement mechanisms to compel major powers like the U.S. and China to comply with international edicts [cite: 56, 57]. The ongoing UN Global Dialogue on AI Governance serves as the ultimate test of whether the multilateral system can establish legitimate oversight [cite: 19]. If it fails to produce concrete, binding outcomes, governance will inevitably splinter along geopolitical and economic lines, with rules dictated solely by those with the greatest technological capacity [cite: 19].

## Bottom line

By 2030, the dream of a single, universally accepted global law for artificial intelligence will be replaced by a complex, multi-polar reality. The European Union, the United States, China, and the newly energized Global South are currently engineering divergent regulatory systems that prioritize vastly different blends of human rights, market dominance, and state sovereignty. What remains deeply uncertain is whether these regional blocs can establish baseline diplomatic interoperability, or if the global economy will fracture into isolated AI spheres defined by digital protectionism. For organizations operating today, innovation is no longer an excuse to bypass risk management; they must adapt immediately to the strict accountability, transparency, and data sovereignty mandates that are already redefining the global market.

## Sources

1. [Global South steps up AI governance following US’s ambitious AI Action Plan](https://dataforpolicy.org/global-south-steps-up-ai-governance-following-uss-ambitious-ai-action-plan/)
2. [Policy Brief No. 225 - Maheshwari](https://www.cigionline.org/documents/3748/Policy_Brief_No._225_Maheshwari.pdf)
3. [Global AI Governance in 2025](https://blog.worldsummit.ai/global-ai-governance-in-2025)
4. [Global South AI Law and Governance Dialogue 2026 Establishes Structured Legal Framework](https://digitalterminal.in/trending/global-south-ai-law-and-governance-dialogue-2026-establishes-structured-legal-framework-for-ai-governance)
5. [The EU AI Act will have global impact, but a limited “Brussels effect”](https://www.brookings.edu/articles/the-eu-ai-act-will-have-global-impact-but-a-limited-brussels-effect/)
6. [Brussels effect or experimentalism?](https://policyreview.info/articles/analysis/brussels-effect-or-experimentalism)
7. [The Brussels Side-Effect: How the AI Act Can Reduce the Global Reach of EU Policy](https://www.researchgate.net/publication/378325303_The_Brussels_Side-Effect_How_the_AI_Act_Can_Reduce_the_Global_Reach_of_EU_Policy)
8. [The Brussels Effect Goes Digital: How Europe's AI Act Will Reshape Global Technology](https://medium.com/@marti.serban/the-brussels-effect-goes-digital-how-europes-ai-act-will-reshape-global-technology-92307d996ff2)
9. [The Brussels Effect: Impact of the EU's AI Act in the EU and Beyond](https://techinsights.linklaters.com/post/102jcir/the-brussels-effect-impact-of-the-eus-ai-act-in-the-eu-and-beyond)
10. [Governing AI for humanity: final report of the UN Advisory Body on Artificial Intelligence](https://www.julia-project.eu/database/soft-law/283)
11. [Governing AI for Humanity: UN Report Proposes Global Framework](https://www.techuk.org/resource/governing-ai-for-humanity-un-report-proposes-global-framework-for-ai-oversight.html)
12. [UN AI Advisory Body](https://www.un.org/en/ai-advisory-body)
13. [The Final Report of HLAB-AI: our analysis and thoughts](https://www.gp-digital.org/news/the-final-report-of-hlab-ai-our-analysis-and-thoughts/)
14. [Globethics Repository: Governing AI for humanity](https://repository.globethics.net/entities/publication/01334166-f983-4257-9151-1312975cb7ff)
15. [The World Is Trying to Govern AI. The UN Wants In.](https://www.cfr.org/articles/the-world-is-trying-to-govern-ai-the-un-wants-in)
16. [What were the outcomes of the Paris AI Action Summit?](https://www.techuk.org/resource/what-were-the-outcomes-of-the-paris-ai-action-summit.html)
17. [The Algorithm Edit: AI Safety Summits](https://thealgorithmedit.com/2025/02/22/ai-saftey-summits/)
18. [What Comes After the Paris AI Summit?](https://www.rusi.org/explore-our-research/publications/commentary/what-comes-after-paris-ai-summit)
19. [Key Takeaways from the AI Seoul Summit 2024](https://accesspartnership.com/opinion/key-takeaways-from-the-ai-seoul-summit-2024/)
20. [Artificial Intelligence Governance to Reinforce the 2030 Agenda](https://securesustain.org/report/artificial-intelligence-governance-to-reinforce-the-2030-agenda-and-leave-no-one-behind/)
21. [UN DESA: Artificial intelligence governance to reinforce the 2030 Agenda](https://publicadministration.desa.un.org/sites/default/files/23rd%20session%20report_Digital%20government%20excerpt_0.pdf)
22. [Should the UN govern global AI?](https://www.brookings.edu/articles/should-the-un-govern-global-ai/)
23. [Government Office for Science published AI 2030 Scenarios report](https://digitalpolicyalert.org/event/29409-government-office-for-science-published-ai-2030-scenarios-report-including-artificial-intelligence-authority-governance-recommendations)
24. [UK GO-Science AI 2030 Scenarios Report](https://assets.publishing.service.gov.uk/media/6808fc002a86d6dfb2b52772/AI_2030_Scenarios_Report.pdf)
25. [CIGI: AI and National Security Scenarios 2030](https://www.cigionline.org/documents/3745/AI_National_Security.pdf)
26. [When AI Meets Climate Action: Five Scenarios](https://ist.psu.edu/news/when-ai-meets-climate-action)
27. [Advanced AI: Possible Futures - CFG](https://cfg.eu/advanced-ai-possible-futures/)
28. [The future of AI interfaces - Research and roadmap to 2030](https://assets.ctfassets.net/3viuren4us1n/3B8hLeUtKqLrPFJf54ZOIg/9c40f75e5d2c86dd52cfbaf100b0de4c/The_future_of_AI_interfaces-Research_and_roadmap_to_2030.pdf)
29. [RAND: Historical Analogies for Technology Governance](https://www.rand.org/pubs/research_reports/RRA3408-1.html)
30. [Exploring Institutions for Global AI Governance](https://aimersociety.com/exploring-institutions-for-global-ai-governance/)
31. [RAND Research Report: What is AI Like?](https://www.rand.org/content/dam/rand/pubs/research_reports/RRA3400/RRA3408-1/RAND_RRA3408-1.pdf)
32. [arXiv: Historical Analogies for AI Governance](https://arxiv.org/pdf/2510.21203)
33. [International Governance of Civilian AI](https://cdn.governance.ai/International_Governance_of_Civilian_AI_OMS.pdf)
34. [African Union Continental AI Strategy July 2024](https://au.int/sites/default/files/documents/44004-doc-EN-_Continental_AI_Strategy_July_2024.pdf)
35. [Emerging AI Governance in Africa: The Continental Response](https://www.trust.org/toolkit/part-2-emerging-ai-governance-in-africa/the-continental-response/)
36. [The African Union’s Continental AI Strategy](https://fpf.org/blog/the-african-unions-continental-ai-strategy-data-protection-and-governance-laws-set-to-play-a-key-role-in-ai-regulation/)
37. [The African Union AI policy framework sets standards for data sovereignty](https://africa.businessinsider.com/local/lifestyle/the-african-union-ai-policy-framework-sets-standards-for-data-sovereignty/n5dpnff)
38. [AU: Continental Artificial Intelligence Strategy](https://au.int/en/documents/20240809/continental-artificial-intelligence-strategy)
39. [Google Search: time in United Arab Emirates](https://www.google.com/search?q=time+in+United+Arab+Emirates)
40. [Google Search: time in India](https://www.google.com/search?q=time+in+India)
41. [Google Search: time in China](https://www.google.com/search?q=time+in+China)
42. [The G42 Enigma: Inside the UAE's AI Champion](https://uecn.org/the-g42-enigma-inside-the-uaes-ai-champion-and-its-global-partnerships/)
43. [The UAE National Strategy for Artificial Intelligence 2031](https://dig.watch/resource/the-uae-national-strategy-for-artificial-intelligence-2031)
44. [Abu Dhabi Government Digital Strategy](https://www.dge.gov.ae/en/news/adg-digital-strategy)
45. [UAE's International Stance on Artificial Intelligence Policy](https://uaelegislation.gov.ae/en/policy/details/uae-s-international-stance-on-artificial-intelligence-policy)
46. [Strategic AI Sovereignty: How Abu Dhabi is Rewriting the Rules](https://aign.global/ai-governance-insights/aign-global/strategic-ai-sovereignty-how-abu-dhabi-is-rewriting-the-rules-of-global-governance/)
47. [Apple Intelligence will be blocked for almost two billion people](https://www.techradar.com/phones/iphone/apple-intelligence-will-be-blocked-for-almost-two-billion-people-and-we-dont-know-for-how-long)
48. [How is Apple working on these AI restrictions in the EU and China?](https://medium.com/macoclock/how-is-apple-working-on-these-ai-restrictions-in-the-eu-and-china-c4430d5dfc4e)
49. [PSA: You can't use Apple Intelligence on devices purchased in China](https://9to5mac.com/2025/04/02/psa-you-cant-use-apple-intelligence-on-devices-purchased-in-china/)
50. [Apple Intelligence EU Rollout Paused](https://asquaresolution.com/blog/apple-intelligence-eu-rollout-paused-future-of-ai-on-iphone/)
51. [Apple Intelligence faces setbacks in Europe and China](https://the-decoder.com/apple-intelligence-faces-setbacks-in-europe-and-china/)
52. [Digital India Act 2025](https://theprayasindia.com/digital-india-act-2025/)
53. [Generative AI Regulations in India](https://americanchase.com/generative-ai-regulations-india/)
54. [Digital India Act - Delhi Law Academy](https://www.delhilawacademy.com/digital-india-act/)
55. [AI Laws and Regulations in India as of 2026](https://www.prashantmali.com/cyber-law-blog-india/ai-laws-and-regulations-in-india-as-of-2026)
56. [Journal of Legal Research: Digital India Act 2025](https://jlrjs.com/wp-content/uploads/2025/08/82.-Sapna-Kumari.pdf)
57. [UN AI Advisory Body final report 'Governing AI for Humanity'](https://www.un.org/en/ai-advisory-body)
58. [CFG: Five scenarios for the AI-transition](https://cfg.eu/advanced-ai-possible-futures/)
59. [UK GO-Science: Five scenarios for AI 2030](https://assets.publishing.service.gov.uk/media/6808fc002a86d6dfb2b52772/AI_2030_Scenarios_Report.pdf)
60. [New Delhi Compact on Artificial Intelligence Accountability](https://www.youtube.com/watch?v=Ycofze7n-Ag)
61. [New Delhi Declaration on AI Impact 2026](https://spmiasacademy.com/currentaffairs/new-delhi-declaration-on-ai-impact-2026/)
62. [New Delhi Declaration Unites 89 Countries](https://babl.ai/new-delhi-declaration-unites-89-countries-behind-global-vision-for-ai-for-all/)
63. [Drishti IAS: New Delhi Declaration on AI Impact](https://www.drishtiias.com/daily-updates/daily-news-analysis/new-delhi-declaration-on-ai-impact)
64. [US Backs New Delhi AI Pact](https://d3f5buclv39urb.cloudfront.net/articles/us-backs-new-delhi-ai-pact-emphasizing-innovation-voluntary-safeguards/)
65. [AI Summit in India concludes with 86 countries signing New Delhi Declaration](https://www.hindustantimes.com/india-news/ai-summit-in-india-concludes-with-86-countries-signing-new-delhi-declaration-101771715565894.html)
66. [AI Impact Summit 2026 Ends With New Delhi Declaration](https://www.timesnownews.com/technology-science/ai-impact-summit-2026-ends-with-new-delhi-declaration-backed-by-88-countries-article-153662153)
67. [TNPSC Thervu Pettagam: February 2026 AI Summit Summary](https://www.tnpscthervupettagam.com/assets/home/media/general/doc/February_25_-_English_Final.pdf)
68. [Current Affairs Notes: February 2026 - AI Summit](https://sathee.iitk.ac.in/sathee-ssc/student-corner/current-affairs/current-affair-notes/february-2026/current-affairs-23-february-2026/)
69. [KGS BPSC CA Magazine - March 2026](https://www.scribd.com/document/1019584209/KGS-BPSC-CA-Magazine-English-March-2026)
70. [Google Search: time in Delhi, IN](https://www.google.com/search?q=time+in+Delhi,+IN)
71. [8 Myths That Are Sabotaging Modern AI Governance](https://www.weforum.org/stories/2026/02/8-myths-that-are-sabotaging-modern-ai-governance/)
72. [5 Reasons You Cannot Afford to Ignore AI Governance in 2026](https://compliancert.com/articles/5-reasons-you-cannot-afford-to-ignore-ai-governance-in-2026/)
73. [AI Regulation in 2026 is Breaking Apart Globally](https://www.questa-ai.com/privacy-cafe/ai-regulation-in-2026-is-breaking-apart-globally)
74. [AI Risk in 2026: When Innovation Stops Being a Valid Excuse](https://www.piranirisk.com/blog/ai-risk-in-2026-when-innovation-stops-being-a-valid-excuse)
75. [Where AI Regulation is Heading in 2026: A Global Outlook](https://www.onetrust.com/blog/where-ai-regulation-is-heading-in-2026-a-global-outlook/)
76. [GO-Science: Evidence and Uncertainties](https://assets.publishing.service.gov.uk/media/6808fc002a86d6dfb2b52772/AI_2030_Scenarios_Report.pdf)
77. [UN Final Report PDF: Governing AI for Humanity](https://www.un.org/sites/un2.un.org/files/governing_ai_for_humanity_final_report_en.pdf)
78. [Ethical AI Design News Release](https://eloncdn.blob.core.windows.net/eu3/sites/964/2021/06/News-ReleaseEthical-AI-Design-6.16.21.pdf)
79. [Futures4Europe: AI 2030 Scenarios](https://www.futures4europe.eu/post/ai-2030-scenarios-gq9lg)
80. [CFR: The World Is Trying to Govern AI](https://www.cfr.org/articles/the-world-is-trying-to-govern-ai-the-un-wants-in)
81. [The Fractured Map of AI Governance](https://whatifai.org/all-insights/the-fractured-map-of-ai-governance-how-the-world-is-dividing-over-ai-laws)
82. [Global Fragmentation of AI Governance](https://bisi.org.uk/reports/global-fragmentation-of-ai-governance)
83. [Mapping global AI governance: a nascent regime in a fragmented landscape](https://www.researchgate.net/publication/353969298_Mapping_global_AI_governance_a_nascent_regime_in_a_fragmented_landscape)
84. [IFOW: Mapping global AI governance](https://www.ifow.org/knowledge-hub-items/mapping-global-ai-governance-a-nascent-regime-in-a-fragmented-landscape)
85. [Mapping the AI Governance Landscape: April 2026 Update](https://cset.georgetown.edu/article/mapping-the-ai-governance-landscape-april-2026-update/)

**Sources:**
1. [questa-ai.com](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQHcfY8ymhX5sZHiZ6Xxhom0m5nfKPkBhFcXEtmJ99bASwDadtRwsZMeczcRaBn42KYPjNnKFtEt6-9-JW0ZvwUWk2Z7_pwUhb2-X1DCznHdk1xX66CvChDhlWRFfsDnpeKTNbrGB-Vev48QkNOAmNI-U3kUFvSobQ4HLHUVzD7QrSNu7U-XI1hnfPezEQ==)
2. [brookings.edu](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQH7fsqqtmJpzVtQ_a6T-GowYQlr97DqMfc-esd15_P3JiKp2i_3euhM9Hu3WszQCjJkY7aUc_z512dXUEu9KIVZTd_D9UIEKpXAY-SWPyEfZcCZxAd9DHBl9UkMbLY3aa4AOWWdbaslmAi-4zprhgy1Go9yGxf5neHNUgvwxT6nV-AW4DGFiuw4TfbyU8yWiMiPvkyfV2Z9rEj6GAo=)
3. [medium.com](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQGQnL1OollSZDtH6HT5K1IVJqMcXD0J88Ht2UXRPscdHOpgoG1fnkb6YTXttfdOuWc4gxHImTK6ZDx_Mq9pRy10okEXT_rQ1eDURoYUBDfXW45ORdPZL0v6K74QgudZYva66II8BqhMwtbW7HHyK2SnMlkSGzmI8fQEzX51no7KVj6fVnp5huMQpNT86h71ssJGwg_5EwaF45CpPrdXI7_gKjZp_8STXk56lABOjvCJ6juxsT89lg==)
4. [whatifai.org](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQEkzcFx0zQpBUi3lrM3Nih-6qKiliv3lINlDyhelTz0sPcDmW5IbjTsciYAyZXDWeboDnVWhsHf4ak4YHUcN4vRATJid4_UVtKkwVOSQZ7T7gK0uxjy0vO7x2tnhyhVACzRWSMaJvNLEvqIbHu3wxaCKJtTvxoj6Nh3G8Gx2fibbg0NMECsMekUGcAwP-cq9XvmXTXy7-90VfhNjnUQ1m2o)
5. [techradar.com](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQFnbW1Zzk-baWxf7ZW00LJASQ97rLzAJuPCMEtui3V4AkMNb7FblB0shFRXXCIVxWJFwJM7rpJek6IYLfJqxOioFqjNWMLs-tUz-eKWVCiTs9ATQSuLpYalrDcxBUMFD38XBOQ9S-qL97PXaOX9wqzzzL3y3q4OR-FHqLPwFTJpkkMvPIxcR2VoyxEzS2O__Jrda0BgNtZ4B1memjIGL7hCWUm02sfWror1NlZLYLkQF84VDLkOVVDVblx4CA==)
6. [9to5mac.com](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQFM2PH0SqAqFmx9iWCFllb4IhwypJx6eXg-6a8gM-V-Yts1VLX2T1evZzkL5FvOyE6hv0Y3m_j2_-QBWtfyOYQJvNnRa379_MIe8NsrNbIdlHVQ7Uki3unrQxnZJNgvdrauoUpH9mSyBmTJBUqklk4u_LwYQ-tOzkSN5kRwNMsRMXUgmMek1-jmFtyeDkfZkNLxdeZfhyg=)
7. [asquaresolution.com](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQFtF6xA0PoROwgtTP_dENY3ZTxLyWkHgwNPf_E8gojKWTXNtSSUW-_SUktwEKblRdP8YGHKWtJqL6JACH_qm4Cq4eyJfWWmWs2P7CSoS7bXf8WwuFQlBC-LKlbSsXjppKR--JS_nSYn4byDgvKm4EZAcGYHyFjIJ7kbHbgDN-foKPwfoe0DFR8LLj18dsndgBwchA==)
8. [the-decoder.com](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQFypfXqNef39SNAsAdDu7G3aUtA-CMIRHKJwkhSBpWQ0vcnyXcK-Ru8_3EescdNo43vmLd1-ftWJeR8QR_aard_HMKlMuvc4UjySkETonUEmTC90RwYZb3iq6Ea9h3iqjtOBDMdxSdLQqqI6cZh73dzaRLplF_p5r9_tNMPKf9bx_mtzQ==)
9. [bisi.org.uk](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQHz8fb-qziUll-1wvG161hhZHcKh-Rk6aKxqbamlVGQxl7fMcPqO8Or-Nmqq2LhK8wmLk-O2byluYaNSvNZZ_FEIBPnakviA3zn7OF1WFg-Dz7FZHYY3H0shjzv0QFCBaHdKP8nWp63DkjGuMuwad9_t9Zqb7wX)
10. [piranirisk.com](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQFBLRyAgioZlxrl9HwyKXYOX_0qReaC0TD4l1LbHm0zGB6mfVHRgPHRB1Vl5qlRBT8g6k6xyLWgpZJdoeo1f9Bcw-ArIhVapKDKH22DvL4ujx9a1bMAlFCTncwnB84tLwAFYa1EezfFgSWwdFEP7RTzjvJ_7fNjBo_MegHEetStuW-0uAls5pnN0eU0hYBuGw==)
11. [compliancert.com](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQHRDX0_oJMIhiOIrC99pKxoNsx83qVIM5kpziNvC-TtIs-PGGDmt9lgkniilK8N36u6lfixouyTEJXcc9_dpUfZQoL0d2NIB-Nqcuqde2BfrICBUZMaPxw246S5LZBGf2T-PE-qAgkwrx2SDhBB_UESNCzzfDHaBurNHyKpsr5uExpsSBsL0jnG-Ju0_ZcFgaWm2_w=)
12. [weforum.org](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQEGRH_4jYi-ghqNHQyWOS4_If3SNKOZ_3uSXianr2P5cRji1XAcsJEUr3L2QgQKGDdAkf0cZfJgR38dz5DMNBiAP-d-chk_Qdxo50nfGt54AwuiueBsrb9b7i1wCpRhnDyFv-dnNw6Ui2tMX-y3TsT4eH4HyJTZLO8sSzan5Jq6i-poCw_u0MpbgNu99vVA)
13. [worldsummit.ai](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQE0SokrDVqzu80tPRzaozzZJONgHGcIvVZBPV-Oo0SiXDDWZc2-cbSetU72-ILEc9LvgspPkpONT7comqfTLutJN1zpCtWEAsDxyL-46naasQ-F-9kYAQiAmf0DeeuDTQ2GUgPv6Wn_mabH8_ep)
14. [linklaters.com](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQG43hdzRqyuEMyjx5YUMEYsytchGfZyZ5o7K6eiRS2qfP_MNqAxi-CfSBNQfhfW1BzIthcW9PKo-2YACTELqdCtsCN7yQRQNWS0l9dQ_INqFaWZYs5AXTQ485k8Ai4SnAA1vkzLdkYufXNx9rYHFsRlqDycuWrwfaKbUcSlyeUucUF3vskJ_5gBcVH1nwTem3MNfI8WtynhdBzAr0iwem1WkyWfcrREeQ==)
15. [researchgate.net](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQHFhHxIRQ_NeZL_0h3Ftz5DWPsV-vFB99AUjjQ8bz5aEIslnDN_qQfF9KkgCghEDu5INqyH_4VaDWWNvGrVPn620GwIUqpV_nZv-ITsPCmkqliWsqTnuKm5axMOTMibjfbnUGvNDt8fMtkKbCaKTw6MvtqGnYTlpvuXcm6xqTyJ2UAwXvW9c2qGOo7JlvgKj0eBGopU1eOsmvIQajOtRE1XYAKEZvHQezCgdoNsgIKQwxa8vkdZ-wY9zw==)
16. [cigionline.org](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQFmY2xmSL7HtOtjQEbGB1ctcUMvj52deOzszmL9e78Y3eG9py2HPVH7MIYmW0ewmMrcPHCFfalHWRm5zpw3el8M9240VjKn45jaF1UUMOEws9AuvO4joV9J_myX9_SI9UxX5Yl5eCfdO0nFkIJhbZURmIzHc5tLad-CBpe7vLppGuyZ)
17. [dataforpolicy.org](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQHhzgEXiYU_0Hx7imJX6BTCYyPkdhRVU8wgc3QTIRp7zw1QgeA5rYVKQR7cpp-jcOoskXfdv3X25A-swq22WxHGRn5KkHDbUudIOJzs79ndQg7b697AzIEbPqXxT69niIpFV5keTS97w0sUrVDBUwFVq8Oc5rzrAYNjOEYjzdNpqO6GUD4mQRV1oqdraNcysFoDHZXmfZ0rz60u)
18. [onetrust.com](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQFc5ST07OK6PiM8u0zTmm5yVUZHjpb7bNRz0eq-FoGVMThLxx22XibiiUdIe4AZIfV2bUfhmmTwxcv-MaEw3K6UOJCWPA-9kr_m_26EnQmGf7638GGTUEpj9qq0po3BDno5a4Znk7xGgI4uhzX7R8LtdzcZnd4zJXCyt2TvhsmyWJ6q4hNX9kKQK3F6)
19. [cfr.org](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQEn2V4pJ1ZzlRTNyNSdEtU6-x75xluP9Lp2V5wMx-kAFWx0rMx30kUe94Dwicjes6YaL-evK6dxONc7AL69GxSiBoQai6Nm-AWK--KcKTdr93dNr-LDPxOjByeCaCC7rYsWkhOpIsG9vcRYvOGY5181pxXn3WsXldlNS-Q-K7yqRa2q)
20. [prashantmali.com](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQEChp64GuNwNAtyV7PuycN_lmvQlGRSFQrWzUp4ClMQooiYQrPq6fafVhD_gDgWPq59g_W11UN7xZqwN0lj0Jf-0WaNtAi_LBQXV6GdC-f5lflNjkas7TnJnfuFptOtZmfdP2PnsQcUjgRj6_49cQnDtmQsofRiD3aFnYL6aEqf1x7Zu0-B-nlfKZGvFUP-G0vDkQ==)
21. [theprayasindia.com](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQH-OTMfZFWp9OeGTXLBb4aiQ1Gj3AzB9EOMDYBHn6UYJW8RL9xKKH4E9MZvyr5G-KfDPAY1ld2Bqgw9cVsvkwLrdh4kl4BfVeC1wZSxt7oaRv_9h5d-x43ESRrdpMq8MJ9YwuRhTHkL)
22. [delhilawacademy.com](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQGl2jZOtElNHbmjBu3VdGl5iGWougA4mx18AJDBIHBGuzB-ZIk7Yvi7AGB5mPD77nP08_qZ4awtw-H46AkldPAibDyVd_xSuRGg2NmfQuiqx9ocOHP2P_NbSYiKvC3jg-ydCqNE8UIw)
23. [americanchase.com](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQGx4DYN4_dnMjQ5GEER9xlq7FZc_ke1zouNmIA-DL0X9Y6MH9bDeVHMhcHa8pbW-uzRAI_dIDadStraPe2FqTw1e_if2EeIudGq4tk-w_O4GHbLPzISxuCzeZEzUAGiKD0tZDiKw6cJ07Rw4FcO5tY=)
24. [thealgorithmedit.com](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQGdW7QkyCfpuExSFn_wGiNK1Ble9wTPqQE2CWfcAlfqGvdbVVYiTtqQ4wt0tKD_eF07W2GbrP6jIbGXNvAQdlUmPtZ0h77-oTJwC-L5qidmJCP_Eb63_JGosfS18eHdVQbiasIdI7k9PIvP7itP0TQ=)
25. [accesspartnership.com](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQGDNnNev-WchtqH5ZcMA-p6qtyzRGDmu3424ILTIPdYKKHJLiWJyLXiPOHS2REuNWR0EeDL7FqZAJuDvIwgdTadaCvti4E3A00SV-UhF0Mq9ZvWDq_1713aADizj32YpOgggY1UdEcR76yfQDpZQO5RotP00NATq60KV6CUl2GE_iIV4ik3e6Y=)
26. [techuk.org](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQHdHkTGeQezEkvgthCoDC_LhwQwHjoPycZ7_WlJIict8VX0HYU7e0rzq1rYcrW-S7n3NsjE_loNpFpoWmLNA1EA44x33_Q5PEvbYkS9A1Z1-XojNk8X7ju98OEhtN8672PkOb_tp0z_wVQTMRni3uIge2tG9f_wU-30rGEF0mlfk6JnHzw682ouAXKhsOiM)
27. [rusi.org](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQGMMpFQO3irQ5ZrotMw3jpnoUJWdunkdZNBAgQvf3puItmUDn3HqYOkGlKptpOmin6zflka_EpnsH92JVBn189K47vxhKY2PiAkf6Ul8S9eUftc73QLah4KnqXysnJxHZBpxTNul462ajleifv4-VomeyNDTvxXRtT5q0xjCHCPe9aTDCxFRqaNff8GDRzSoazNIFvUZw2a)
28. [hindustantimes.com](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQEjven5A9oyh-yOhPl9LE95kBuKq51jypjWtYa-iV9dzoFk72vRnDtsX_HapIGvwwgME0ZRfmTtpT8TI0iH6pyynO6RKdCTJwNTMffd-J4_84x4D0ATxoFE96KQ6mr1-b0d5EogTdBh2Zauq3Ql7zt2S694VInzqEVWsba7ci5VNnhOFG2_9WKRQg1pjTeo8qVBG4I8DkC4YRIBa7-JKAEhcJ-Ae1qWPeZcZK1GrTZ7M0p96OmKyjQgAPK5ddTDWdIN)
29. [spmiasacademy.com](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQEJVl44pNIM2o0itYXgLGWYo-vU2-s_fcIZ9QBgwimjgXxZU9EOIGe7Q2CgvgEfA8YxUlgVSw9HgPeiLxZPpghx6tkITp4OXexXdCrulCzhPJwlRNVmPsK98oOJIRMbVEpsZa7HnvxKxS4v4Osr8suRwtEBssFEREVJ22ZFQ5tG7VOaQlilIw==)
30. [cloudfront.net](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQEQyUxf_jEOCLCXviJpfHtxHfth5m7sYADXk_oR5WeuUQe-pDxQ2NZ9wRv_QlhfKnj5xELRpIpiMWkpDe9FAi9J1JXPbBRcMcGvNdANrl-bRwP0KF5Y3NI15LhPr0zG2cfC10BXromTevJcYT0_5h2QCnMtRCuMZww-WP_BALL_VrS8txVHmcLflWdvG8oVkcQoV2FmoNOaoHEHEx_7v0mNitLgBa7TjZGynOg=)
31. [timesnownews.com](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQGbOi3xB1wnO4L_UDsO8gndtxTo8Hb-Kgc5VKC44sQGfjg9E0fruNJGgrMuEP1dRdOwAINBVzDuLdHsVdowDZjJBh-WTtCEt1pAW30wXDIiJoxHnRQcZypaaip1cXLp71PT_MfMow7JDodV7aI8d3R-pdG7bjI_Uid2MOkvyPxIXQbWUB_F-5e32hJrT1ekvZj2zb0bdyHGWRfQ80KQt7J8G5cZZouOh_1MVSyV0OEv5rv_1Bl2tp8ppyWybAvp4CDTzCM2)
32. [drishtiias.com](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQFNX2KD8h1vuBL22Bz9kHGyVrbRQDPz3ozDGwMS6HvcY8HJkGrEFYv0GuSjwv32rnwvQYUlbxB3UtFk3h6WRvtXcZiszhZ3XrQXHGhdTzd5x9386SH6NAm0WkFHaqxHSi0WNuc4loowMnhn3eqvEsAY4tG80EzqOqMzEU_CeT8AHrEfAFS5eVF1ijeXFzR3fLjTf7Xm)
33. [babl.ai](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQGVbExBItqGJFlFF-BDI6G6s-ycGGb224a4c0_mijMqgypLZbtDHD0cTowLJsDucKS1lHUw4SgTnryyUFisvB09gSPumRUs8AcLpz4GgarkuLMPyTsne4-9N9Wm02V5rloqdKPeikCRaTQa6sX45xSVasLQ76gsXEUlsJffAflcOxE4RaQE1VJq9cLyTBHUYBMx8-Y=)
34. [youtube.com](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQFAslfztapNXvi1XEKOk22ClVW1X_S5bNiQNsqFEyWMsLOd388cTNX9ZuZpqnLX6byPRdJua_08jU1CRaWJL-tVAy3RyTdVM4ri0sS6gJrfhBg21-h3RqkGVPoA75SstxQ=)
35. [digitalterminal.in](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQE14LFdleF1xn5X-xTo8BiAomK7MFDaY6g4YYPVSo315SrR91InQENXhgWNqgol-6BYpFIkLlgZVczZG70vjIa90drb-ggrN1sNtHqN10sl5lLnLCiKewRAWuWUdiEYYDoiQKktQnsoQ38tqXfjZPlBzLxwkfWqrupLR5h3Rmi5LUkBxI31XmhACn6b3mf_fSxqoJAwwb-QmTYX9_LsrwSR6uSsaNI_Ci8YIEUq8VUuRRDIH042bspX9Fbs-j_jrtcFZ7Y=)
36. [au.int](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQF5QnDvPfU3zTpXgpDu4EKPD_sHZ75pPr_0VtHTSlk0tSw9E-hxgEutlnWjcal5uTldbFL4-fhhDi8M8L8VxMujThIgGt5HgC5Biy-hyP_2CNvTHq6cOnJ9FdNVo7BVdXeozMZbzAtH-Y5EfLZhggwabGOLbzvqrphV03geYcaE5lDPzaBOWxAwe3nSKbuh4zDy0c10gA==)
37. [au.int](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQHP2JZLHMHXIHenloegVS8q0FyWvlo1XMHbNNO2PIR1blOj0T5xxuFgr2wzQLNBBZJaCRrmRmgutB9TU1dmTzOqe0PoCFfNqIWmZ82942tQiJke0LabBF7SzWej2ZtshWetvVG5eUa4_w2zYYUbH9BhDqvApJAg3GV5CT6C4qn303os1VNjxw==)
38. [businessinsider.com](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQHfFqFBZnFOedJ_EXNUJ0uAtDJbZDGuacvSxwsWj5h3XTRtsZvbJR7fLaElZU3k8qUsyYpwRFMLvB2oYqwCbtU6Wl4SoQHIZQWbUIyvVqxPwwFH6M_-O1f4LZamOBvj6bW46qSu1I7zr-8zA0xwo8Ol5G7DDg1a9oglfLrbu5RygUEbhTc0WEpqECBCtGWxfwDajxFl6sMZxtqd4nauYv-eCCm_Z9LQ66-QgqhX5OukFy_CMv5b6LplbY4=)
39. [trust.org](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQHA2-a8-Z8rwWQ3utkbdA26fK6m_1xSFni69GO32x7-F9bpv-Hz_Tjhstl9rBUOZbZGCRE0kLliLO1IXm77lBJvgrSfvJ9cNO4R5v3Td7D6KRphfT-EBj5yVmwHDrkwPHngchpLJgQ7_3_RnN2GNhZGXXRbhi1T8E5FTmDYSe5-uCTvVlSCGozwWd-tG6MCn-Ah4rSp)
40. [fpf.org](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQG3-k6D2hlibEuJ2zwiYcue-5eJY1g5cJcGpuBGe5VI3R2rQlM3DYRZS299ORFrBgZbzonpRAmippfFxq8Sp4mZtl1VhflmAnjSsBaHqVrF7nghXR10M7XZyC98Ua02sjVQ_IrUmXfgocov8xWKQwmaNUg3EAPZqj5D_rTqVW8DzkHLpKFsLh8hJHd6o6OJRSEQ5XSzgCSzlpDqW9kANpziwuKwM6E-MjQQa5dHpMR-KDlIXxN5iYcCi7C2KHhv8EtcQQ==)
41. [dig.watch](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQFDRvF7DQQgiCTrzR8jsSu6EuqCmYCbf2IbaQqZOqB_NoxZUY_TTg8315Jd6PTMfUImjs-m2-K6KVmzSHVr2Au1_UaGhH8lflIwZpos6nWxSp60tSn_9LnjQf3dPSdcP94fTivX2jChxZtPZ6Se46xXsWQ-cFpVVYaIIEdyCOq0s7DC8CzjPYofvy4=)
42. [aign.global](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQFMLxin8_LyS8a3fomc0rl8rcpjppOJ7Q1xOnzm4bfavizUwgTVgQRwyu--5KfrL31QgVFNK4a_vAGVJRjXkODpDydOdEBsIUypkNy3Ft8DeX0QOMYYUP89rJI1iCm6lG2TYP_gctuoncppNG12ZaHlG1ucYXXVOQ12XcsPCUR_U8pSoSurNK1P5u07RqMh-t6PaJkjw26B1_iNPB98K-NhWQDemExjfIMv2pogMaE730W7s7zSuln0Ozp3UgncuzM=)
43. [dge.gov.ae](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQGwNrNIOe6QkI70nhAtUR7XNPlq8fgQUKOB5zDc41cRT37OUFaK-6iAetFlugpnW3W6bPG0JhhKy76rqPwOB_VE19hzAmuwca9Xncb_LF5lT0vhsPAMqTKTFL4In9171Lkjn7fityRnOw==)
44. [uecn.org](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQFqY0IQvuvW29YHIDtF8zEEPCU5EgddL-mUwosesybbi5BwgFL5bPWetMH77lorLKMh9Xn6_DPuOGDdCIrTG53AUeuVFD1K7hAIRRMnZ2EPFu2hsDrot81CtJ2vYnb0gzrfM4MKwxMCEBQ8WxjFPbXgZ5a3UggEfSbUpv2bZeP6jxjGhhSjA0euGqlVgQI=)
45. [service.gov.uk](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQHZPGEDfcUc0O9isXr399r96eQM3FUZAbmZ4NYWghTs7gIFcz6MSvzlmreSCW9oerUCzgrqMaRaR8UvvXmmaiclsbS8i074rq6A6MT1qkuYsQ6XfV5O0yNWKIqPHThyQPN5oBFq8d960hDDSyCzXWaA5mXa6zQoL5ALrXiUvBJImCp2T65HFnaSDp6ap_JkhXuAPgiHooP09Ok=)
46. [cfg.eu](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQFlA3aNdBf9V5IksWWtFAoaZQXOP2KyXbY7JQ9taMTDQr9PfXFFeQZ_Ox6L9xeHOmF1CV0DwRP0et9uloWwbmDviB_N_03hrYvFR9koQhfk0p-AClNB-51m6Oj-l---vg-1)
47. [futures4europe.eu](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQGr82qFiNtlVaWaJDAiNKadlvhRlj9XFYDHMtkFU6AgVc6tLNhrEyFBQ-PGwEvny7hAnldi7p_ooG3Uz0XlyUpID5ByKoDgSW0pBPXRA0Ue9eHAmfR2Ega7tc3Yq030ZQ4iCzA2Vsk-nSQlZqaSrxQ=)
48. [techuk.org](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQGN0oB10REQpeVhT2oqEqYHWCUjbfqevuXnJAZncRnRdsD55qcBT1BHJLAHLVpUUCVOfY2oUguDBHtaJZZxLE-jmuRs8uO-FBxuRhiDBgjUXJYgMGL41sASgAIikxywhohsn2q5hWgiWPlecsDrWPQ4AmHL9LcQ2cp0FPVjMM2pxJFtZiexGCl3VMgKgAN_K8buDxa-EnZIbnx6S4nkC_MTPG_RyKpaEXU=)
49. [un.org](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQEkrj1sN3K2X6p9t-TTt_JDJf2V0f_lm8zttvC6seb_WKYT1dxjETwSHd0H4-72jU4p1ukVHXCA6m98rhpr9gm9NimD5Xh7bfh6wW3VLLdgIgmZYK23bTAAhCnVNtL6EiclYuon_8HFoeNz6lnEgma6Cslg_r5p7yxYt8BTUcE5XNM6o6mzhFjz-TCQ0Q==)
50. [rand.org](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQFo7-FUIRCpS9VecdLA_9QUJkv3c06m1Ozjf_Ke8Z10rtiSqcVkC4RoSI1Uzhz9RKaFxzppH7TtByR9NdMU0JBhZ9cqjfj6iC2B042BLCYOdZ-sNsuuvt7fNZyqtVAlCDb627yYjqNPjiUHend67g==)
51. [rand.org](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQGPmggEKWHjpp4LyPTByAFsrV5Q-3MFu3CHTa77PLRGdGRCx1FtU4E4JYsMWVziDp9aogSz_6rt_Ae003jOd5HYr2rVueL1ZAgOtzNHhzkP29nsbq_9EnDVR18tctYcNk4D852zWVKhoLYRH2ZrEDkRUG114wEiXZfrORsPBwK0NsOBljiQ0NvmloKozbZYupWjDh33WA==)
52. [arxiv.org](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQH1hpb8cszACZSwXgTcFAGemIGIw8OppFKPNg5rTSNGaEAOnMaBQpXpnsBLfUtVF8NUH9aJAjjmhhM0PumcAjoO8cQAf3JTw6SSRvFi-Z2Nx_hiU_Qy)
53. [aimersociety.com](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQHkE28N8vTUZT1vY9KFy06P-kRgllrwkMHtILC9snQazXl8jj3tG3ppfbVTvPwqfUtWU1MPllblMv53xUgDgZG9NCau7OK8rrObEGjzaOpe7Ev_UYE5s8YSbrh6KiE7M_B4PVD0NNs3OmSo4PsBcjJuGisW17bgyhdlgvCqc1s=)
54. [governance.ai](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQG5lbdJWDUQJY5VMu3y-teN9mCxFciKTnJC76aN4QWpxzviNT-0TmmnzaYZbEm4IKRcxEh9oLzuUftQETcLU6h55HJcSgw-bKm8RIT7wymFwdCqkTkbnChb2884wZbcYO91uwf7nVOSldYttIZD81TPLB_rRWpu92VIX5pzD80=)
55. [julia-project.eu](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQET_O7x11RI5_lZDoJ-dwxm_w_No96awyWYqM4kuJiCAWaan73OAQ3Q1Gn8F6iqkLRPy3jXp2XIT7ZhF8syh0bR9Xw3swU8cukdrzm7GOWE8QkOPpzQFBz_0YxyZgk0TRVOV4g7QFxf)
56. [gp-digital.org](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQHp8wVQ-lkhFnN0ZhjcrzZaKFEXJroH53S_GVHoHwSL1my3j8mt44FsuShHVcM7__PNCk_C64DPrjfw7B-UBW-HY88k-YFtlyf4TBL43Zbvt3K1EWVZK82OcCVMkUVValG4X53hLWW1SFAgv2CEPEAo4gvtMdHg2rXViTEcdUdpaRAcyyjymoSLTBh_)
57. [brookings.edu](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQGYApgd7dJR2VjbGfID0cSvFXTvk5f50SrLYBgHH6TsF7fAuIgv-GPtKVDJgXurJYYOEN_xOw5eqNsYWNEpyQRvVxiF_sUDcgMVwl8GmBpr7nBc8mJ-fqi829MsHAC18XVG0Y_W_gqtGd9IE1dlgf6sMaVxtD1t1w==)