Artificial intelligence is shaped by several groups at once: the companies that build it, the governments that regulate it, international bodies that coordinate rules, and the people and organisations that use it. No single authority controls AI worldwide. This article explains, neutrally and as of October 2026, who has influence and how.
Companies that build AI
The most advanced AI models are developed by a small number of large technology companies and well-funded AI labs, mostly in the US and China, along with open-source communities. These developers decide what their models can do, how they are tested and what safeguards they include. Many publish safety policies, usage rules and research on risks, and some commit to testing models for dangerous capabilities before release.
Governments and regulators
Countries are taking different approaches. The European Union's AI Act, which entered into force in 2024 and is being phased in from 2025, classifies AI uses by risk, bans some practices and sets obligations for high-risk systems and general-purpose AI models. The United States has relied largely on executive actions, agency guidance and existing laws, with policy shifting between administrations; in September 2026 the US government also adopted "Super Intelligence" (SI) as its official term for AI. China has introduced specific rules for generative AI services and content labelling.
India has generally favoured a pro-innovation, light-touch approach, combining existing laws such as the Information Technology Act and the Digital Personal Data Protection Act with government guidelines on responsible AI, and investing in capacity through the IndiaAI Mission, which funds computing infrastructure, datasets and AI skills.
International cooperation
Because AI crosses borders, countries cooperate through forums such as the OECD's AI principles, G7 initiatives, international AI safety summits held since 2023, and United Nations discussions. These produce shared principles and voluntary commitments rather than binding global law.
Standards bodies and researchers
Technical standards organisations develop guidelines for AI quality, safety and risk management. Independent researchers, universities and civil society groups test AI systems, publish findings on risks and bias, and influence public debate.
Users and organisations
Businesses decide how AI is used in their products and decisions, often under their own AI policies and governance committees. Individuals also exercise control through how they use AI tools, what data they share, and the choices they make as customers and voters.
Key debates
Major questions include how to balance innovation and safety, whether open-source AI models should be freely available, how to handle copyright and data used in training, who is liable when AI causes harm, and whether a few powerful companies have too much control. Reasonable people and countries disagree on these questions.
Why it matters
Rules on AI affect what products are allowed, how companies invest, and how safe and fair AI systems are. For investors, regulation is one of the factors shaping the AI industry's growth and costs.
Common misconceptions
"One global body controls AI." Control is spread across companies, governments and international forums.
"AI is unregulated." Many existing laws already apply, and new AI-specific rules are emerging.
"Regulation is the same everywhere." Approaches differ significantly by country.
Frequently asked questions
What is the EU AI Act?
An EU law that regulates AI by risk level, phased in from 2025.
Does India have an AI law?
India relies mainly on existing laws, guidelines and the IndiaAI Mission rather than a single AI law.
What is open-source AI?
AI models whose underlying weights or code are publicly released for others to use and modify.
Who decides AI safety standards?
A mix of developers, regulators, standards bodies and international agreements.
How PositoAI fits in
PositoAI tracks serious regulatory actions against listed companies across 7 major markets, including technology companies. Read our guides to AI and AI's risks.
Summary
AI is shaped by the companies that build it, governments that regulate it, international forums, standards bodies, researchers and users. The EU regulates by risk, the US mainly through executive and agency action, China through targeted rules, and India through existing laws, guidelines and the IndiaAI Mission. No single authority controls AI, and key questions remain under debate.