The Creation of WAICO, the Dispute Over the Global Governance of Artificial Intelligence, the New Gartner Magic Quadrants, and Brazil’s Delicate Challenge of Regulating Systems That Are Already Redesigning the World’s Economic, Institutional, and Legal Infrastructure By AI CHATGPT with GOOGLE AI

 

The Creation of WAICO, the Dispute Over the Global Governance of Artificial Intelligence, the New Gartner Magic Quadrants, and Brazil’s Delicate Challenge of Regulating Systems That Are Already Redesigning the World’s Economic, Institutional, and Legal Infrastructure

By AI CHATGPT with GOOGLE AI

Introduction

On July 16, 2026, representatives of 29 countries signed, in Shanghai, the Agreement on the Establishment of the World Artificial Intelligence Cooperation Organization, known by the acronym WAICO.

This is not merely another international forum, a diplomatic conference, or a political declaration without a permanent structure.

The agreement formally creates an independent intergovernmental international organization, endowed with international legal personality, headquartered in Shanghai, with a deliberative Council, a permanent Secretariat, its own financing system, the possibility of receiving contributions from governments, international organizations, companies, individuals, and other entities, as well as privileges and immunities to be defined in subsequent agreements.

The birth of WAICO takes place precisely when worldwide investments in artificial intelligence are reaching an unprecedented scale. In May 2026, Gartner updated its forecast for worldwide AI spending to US$2.59 trillion in 2026, a 47% increase compared with the previous year. More than 45% of this spending is expected to be concentrated on the physical and computational infrastructure necessary to operate artificial intelligence: servers, semiconductors, networks, data centers, cloud services, and processing capacity.

The combination of these two events is decisive.

On one side, States are seeking to create international rules, technical standards, cooperation mechanisms, oversight instruments, and safety principles.

On the other, private companies, cloud providers, semiconductor manufacturers, model developers, platform suppliers, and data-center owners control a large part of the material infrastructure upon which artificial-intelligence systems exist.

The central issue, therefore, is not merely which country or international bloc will write the rules.

The deeper question is:

Who will design, own, finance, control, update, audit, and be able to shut down the artificial-intelligence systems that will begin to mediate economic, judicial, administrative, medical, educational, military, and social decisions?

It is at this point that the creation of WAICO, the European AI Act, Bill No. 2,338/2023 in Brazil, the new markets analyzed by Gartner, and the structural approach to AI-system projects converge.

CHAPTER I

WHAT WAS ACTUALLY CREATED IN SHANGHAI

Official Chinese sources confirm that 29 countries signed, on July 16, 2026, the constitutive agreement of WAICO.

According to the Chinese government, the organization will be independent, intergovernmental, headquartered in Shanghai, and committed to the purposes of the Charter of the United Nations, to a human-centered approach, and to the beneficial, safe, and fair development of artificial intelligence.

The full text of the agreement, made available by the Digital Watch Observatory, confirms that WAICO was not conceived merely as a forum for debate.

Article 1 grants the organization international legal personality and full legal capacity.

Article 3 establishes concrete functions, including:

• creating a platform to match supply and demand among Member States;
• promoting international capacity building in artificial intelligence;
• coordinating development strategies;
• bringing governance rules closer together;
• coordinating technical standards;
• promoting interoperability;
• sharing best practices;
• encouraging open-source ecosystems;
• cooperating with the United Nations system;
• maintaining relations with other international organizations.

Article 5 creates two central bodies:

1. the Council;
2. the Secretariat.

The Council will be the political and institutional decision-making body. It will have authority to:

• define the general strategy;
• approve regulations;
• approve the budget and accounts;
• approve annual plans and reports;
• admit new members and observers;
• appoint the Secretary-General;
• create subsidiary bodies and working groups.

Each State will have the right to one vote.

Decisions should preferably be taken by consensus. When consensus cannot be reached, they may be approved by a two-thirds majority of the States present and voting, provided that there is a quorum of at least two-thirds of the members.

This demonstrates that WAICO has a real institutional structure, not merely a symbolic one.

CHAPTER II

THE FULL TEXT OF THE WAICO CONSTITUTIVE AGREEMENT IN PORTUGUESE TRANSLATION

AGREEMENT ON THE ESTABLISHMENT OF THE WORLD ARTIFICIAL INTELLIGENCE COOPERATION ORGANIZATION

The Contracting States to this Agreement,

Recognizing the transformative power of artificial intelligence — AI — as a driving force for economic and social development, capable of accelerating the implementation of the United Nations 2030 Agenda for Sustainable Development and having repercussions on the well-being of people around the world;

Recognizing the need to provide more international public goods and to participate in international cooperation and the global governance of artificial intelligence, in order to promote its sound and orderly development in a beneficial, safe, and fair direction, for the benefit of all humanity;

Desiring to contribute to the purposes and principles of the Charter of the United Nations;

Decide to establish the World Artificial Intelligence Cooperation Organization and agree as follows:

Article 1 — Establishment

1. The World Artificial Intelligence Cooperation Organization, hereinafter referred to as the “Organization,” is hereby established as an independent intergovernmental international organization endowed with international legal personality.

The Organization shall have full legal capacity and shall operate in accordance with the provisions of this Agreement.

2. All Contracting States to this Agreement shall be Member States of the Organization.

Article 2 — Purposes

The purposes of the Organization shall be:

a) to adopt a human-centered approach and promote the innovation, development, deployment, and application of artificial intelligence for good and for all;

b) to expand international cooperation in artificial intelligence and jointly establish an open, equitable, fair, and non-discriminatory environment for its development, with the objective of unlocking the benefits arising from artificial intelligence;

c) to reduce inequality in access to artificial intelligence and promote greater accessibility to AI technologies and services;

d) to mitigate and collectively address the risks, challenges, and impacts arising from the advancement of artificial intelligence and promote AI that is safe, secure, ethical, and trustworthy.

Article 3 — Functions

1. The functions of the Organization shall be:

a) to build a platform to match supply and demand among the Member States;

b) to strengthen international cooperation in capacity building in the field of artificial intelligence;

c) to strengthen the alignment and coordination of artificial-intelligence development strategies, governance rules, and technical standards; facilitate dialogues; build political consensus; promote interoperability; and share best practices;

d) to strengthen scientific and technological cooperation in artificial intelligence and encourage joint efforts to develop an open-source ecosystem;

e) to strengthen the global governance of artificial intelligence, support and promote convergence with processes developed within the framework of the United Nations;

f) to maintain communication with other international organizations and explore appropriate forms of cooperation.

2. The functions of the Organization shall be limited exclusively to the civilian domain of artificial intelligence.

Article 4 — Fundamental Principles

The Organization shall follow the purposes and principles of the Charter of the United Nations, respect the sovereignty of States and the diversity of civilizations, uphold the principle of equality, practice multilateralism, and promote the benefits of artificial intelligence through broad consultation and joint contribution.

Article 5 — Structure

1. The Organization shall establish a Council and a Secretariat.

2. The headquarters of the Organization shall be located in the city of Shanghai, People’s Republic of China.

3. The Organization may establish subsidiary bodies and working groups as necessary for the performance of its functions and the fulfillment of its purposes.

Article 6 — Council

1. The Council shall be the decision-making body responsible for:

• defining the general strategy and regulations of the Organization;
• examining and approving the budget and final accounts;
• approving annual work plans;
• approving annual reports;
• admitting new Member States and observers;
• appointing the Secretary-General and Deputy Secretaries-General;
• creating subsidiary bodies and working groups;
• deciding on other matters presented by the Member States.

The Council shall also exercise such additional powers and functions as the Member States consider necessary for the implementation of this Agreement.

2. Each Member State shall be a member of the Council.

The Council shall hold one plenary session annually.

The President of the Council shall serve a one-year term and shall be elected by the plenary session, taking into account equitable geographical distribution.

3. Each Member State may designate an official coordinator, responsible for maintaining the necessary communication between the State and the Secretariat and for submitting recommendations to the Council.

Meetings of the official coordinators shall be convened under the direction of the Council.

Article 7 — Decision-Making Process

1. The Council shall adopt its own rules of procedure.

2. Decisions shall be taken by the plenary session of the Council, with the presence of at least two-thirds of all Member States.

Decisions shall be adopted by consensus to the greatest extent possible.

3. When consensus cannot be reached, despite all efforts made, decisions may be taken by a two-thirds majority of the Member States present and voting.

4. Each Member State shall have the right to one vote.

5. The rule provided for in paragraph 3 shall not apply to amendments to this Agreement.

Article 8 — Secretariat

1. The Secretariat, as the permanent administrative body of the Organization, shall be responsible for:

• implementing the decisions of the Council;
• preparing the budget and final accounts;
• preparing annual work plans;
• preparing annual reports;
• managing projects;
• coordinating with other relevant international bodies;
• carrying out other tasks assigned by the Council.

The Secretariat shall operate under the responsibility of the Secretary-General.

2. The Secretariat shall have a Secretary-General, who shall be the principal administrative officer of the Organization.

The term shall be three years, renewable once.

The Secretary-General shall be a national of a Member State, nominated by a Member State, and appointed by the Council.

3. The Secretariat shall have two Deputy Secretaries-General.

Their terms shall be three years, renewable once.

They shall be nationals of Member States, recommended by the Secretary-General, and appointed by the Council.

The remaining staff members shall be appointed by the Secretary-General in accordance with the regulations of the Organization.

4. The paramount consideration in the employment of staff shall be to ensure the highest standards of efficiency, competence, and integrity.

Due regard shall also be paid to the importance of recruiting personnel on as wide a geographical basis as possible.

Article 9 — Financing

1. All financial matters relating to the Organization shall be governed by this Agreement and by the financial regulations adopted by the Council.

2. The Organization shall obtain resources through:

• annual contributions from Member States;
• voluntary contributions from governments;
• contributions from international organizations;
• contributions from companies;
• contributions from individuals;
• contributions from other entities;
• lawful revenue generated by cooperation projects.

The Organization may accept contributions only with the authorization of the Council.

It shall not accept contributions or assistance that may prejudice, restrict, distort, or alter its purposes, objectives, or functions.

3. The annual contributions of Member States shall be determined by the Council according to an agreed scale, taking as a reference the contribution classes of the Universal Postal Union system.

The level of economic development and the capacity to pay of the States may also be taken into consideration.

Article 10 — Accession

1. Accession to the Organization shall be open to all States.

2. States that sign the Agreement during the period provided for in Article 14 shall be founding members.

3. Any other State wishing to become a member shall submit an application to the Secretariat and, following approval by the Council, accede to the Agreement under the terms of Article 14.

Article 11 — Observers

1. The Organization may grant observer status to non-member States and intergovernmental organizations.

Interested States and organizations shall submit an application to the Secretariat, subject to approval by the Council.

2. Observers may participate in the plenary session of the Council upon invitation by the President.

All expenses arising from participation shall be borne by the observers themselves.

3. Observers shall not have the right to vote.

Article 12 — Privileges and Immunities

The Organization, its officials, and the representatives of the Member States shall enjoy, in the territories of all Member States, the privileges and immunities necessary for the performance of the functions and the fulfillment of the objectives of the Organization.

The extent of these privileges and immunities shall be defined in separate agreements.

Article 13 — Settlement of Disputes

Any dispute concerning the interpretation or application of this Agreement shall be settled through consultations among the Member States.

Article 14 — Signature and Accession

1. This Agreement shall be open for signature between May 1, 2026, and July 31, 2026.

2. The Agreement shall be subject to ratification, acceptance, or approval by the signatory States.

3. The Agreement shall remain open for accession after the close of the signature period.

Article 15 — Entry into Force

1. This Agreement shall enter into force on the date of deposit of the third instrument of ratification, acceptance, or approval.

2. For each State that ratifies, accepts, approves, or accedes to the Agreement after the deposit of the third instrument, the Agreement shall enter into force on the date on which that State deposits the corresponding instrument.

Article 16 — Amendments

1. Any Member State may propose an amendment to this Agreement by submitting it to the Secretary-General at least two months before the plenary session of the Council.

The Secretary-General shall inform the other Member States of the proposal.

2. The amendment shall be examined and approved by the plenary session of the Council by consensus.

3. The approved amendment shall be subject to ratification, acceptance, or approval by all Member States.

It shall enter into force on the date of deposit of the last instrument of ratification, acceptance, or approval.

Upon its entry into force, it shall replace the original amended article.

Article 17 — Withdrawal

1. A Member State may withdraw from this Agreement and from the Organization by means of written notification to the depositary.

2. The withdrawal shall take effect one year after receipt of the notification, unless the notification indicates a later date.

3. The withdrawal shall not prejudice obligations undertaken by the State before it takes effect.

Article 18 — Depositary

The Government of the People’s Republic of China shall be the depositary of this Agreement and shall notify all Member States and the Secretary-General of:

• the date of entry into force;
• signatures;
• ratifications;
• acceptances;
• approvals;
• accessions;
• other relevant acts.

This Agreement was drawn up in the Arabic, Chinese, English, French, Russian, and Spanish languages, all texts being equally authentic.

The depositary shall transmit certified copies to the governments of the other Member States.

In witness whereof, the undersigned, duly authorized by their respective governments, have signed this Agreement.

Full source: Digital Watch Observatory.

CHAPTER III

WHAT THE AGREEMENT ACTUALLY REVEALS

The full reading of the agreement makes it possible to move beyond simplifying interpretations.

WAICO does not establish only abstract principles of cooperation.

It creates an institutional infrastructure that may participate in defining:

• technical standards;
• governance rules;
• interoperability systems;
• international capacity-building programs;
• open-source ecosystems;
• cooperative projects;
• financing mechanisms;
• coordination with international organizations;
• the production of political consensus on artificial intelligence.

Article 3 is especially important because it places, on the same level, three dimensions that are normally treated separately:

1. development policy;
2. legal governance;
3. technical standardization.

This combination is decisive because, in the field of artificial intelligence, technical standards can produce normative effects equivalent to or greater than those of certain laws.

A legal rule may state that a system must be safe, transparent, or auditable.

But technical standards are what will concretely define:

• what will be considered safe;
• which tests must be performed;
• which records must be maintained;
• how training must be documented;
• how incidents must be classified;
• which metrics will be accepted;
• which interfaces will allow auditing;
• which requirements will be demanded for interoperability;
• how explainability will be measured;
• which human controls will be considered sufficient.

Those who participate in the creation of these standards influence the very operational definition of legality.

CHAPTER IV

PRIVATE FINANCING OF WAICO AND THE PROBLEM OF CORPORATE POWER

One of the most relevant provisions of the agreement is Article 9.

It allows WAICO to receive voluntary contributions not only from governments and international organizations, but also from:

• companies;
• individuals;
• other entities;
• cooperative projects.

The agreement establishes that these contributions will depend on authorization by the Council and may not restrict, distort, or alter the objectives of the organization.

This safeguard is important, but it does not eliminate the structural problem.

Global artificial intelligence is dominated by a limited group of companies that control:

• semiconductors;
• data centers;
• cloud services;
• foundation models;
• development platforms;
• data;
• deployment tools;
• security systems;
• distribution channels;
• interfaces used by governments and companies.

If these same companies may finance projects, capacity building, infrastructure, or programs of the organization, it will be indispensable to create rigorous mechanisms of:

• financial transparency;
• conflict-of-interest disclosure;
• tracking of donations;
• prevention of regulatory capture;
• public access to meetings;
• disclosure of technical documents;
• separation between funders and bodies responsible for drafting standards;
• external auditing;
• civil-society participation;
• control by Member States.

The issue is not to presume that every corporate contribution is illegitimate.

The issue is to recognize that, when companies that will be regulated also possess the material capacity to finance the environment in which standards and policies will be discussed, an objective risk of institutional capture arises.

CHAPTER V

BRAZIL BETWEEN THE EUROPEAN AI ACT AND WAICO

Brazil occupies a particularly complex position.

Its General Data Protection Law was strongly influenced by the European model.

Bill No. 2,338/2023 also adopts an architecture based on:

• risk classification;
• rights of affected persons;
• governance duties;
• impact assessment;
• transparency;
• State supervision;
• accountability;
• sanctions.

At the same time, Brazil joined the Chinese initiative to establish WAICO. In February 2026, before the formal signing, President Luiz Inácio Lula da Silva himself stated that Brazil had joined the Chinese initiative to establish a world organization for cooperation in artificial intelligence, aimed especially at developing countries.

This dual participation is not necessarily contradictory.

The European AI Act and WAICO do not operate on exactly the same plane.

The former is supranational legislation applicable to the European market.

The latter is an intergovernmental organization intended for cooperation, global governance, and the creation of standards and projects.

Nevertheless, tensions may arise when technical standards, interoperability mechanisms, certification regimes, or governance models developed by these different poles are not compatible.

Brazilian companies may be required to maintain:

• different documentation for different markets;
• duplicated testing;
• incompatible risk classifications;
• separate registration systems;
• different auditing procedures;
• divergent transparency standards;
• different human-oversight mechanisms.

This situation may increase compliance costs and favor precisely the largest companies, which are capable of maintaining multinational legal and technical structures.

Smaller companies, universities, civil organizations, and independent projects may be excluded by the cost of regulatory complexity.

CHAPTER VI

IS WAICO AN ORGANIZATION “WITHOUT IDEOLOGICAL RESTRICTIONS”?

The summary produced by AI Mode stated that WAICO does not impose tests of political alignment or adherence to Western democratic values.

This statement finds partial support in the text of the agreement.

Article 10 establishes that accession shall be open to all States.

Article 4 emphasizes:

• sovereignty;
• equality among States;
• diversity of civilizations;
• multilateralism.

The agreement does not contain an express requirement that a candidate State be a democracy, belong to a specific political bloc, or adopt a specific model of rights protection.

The academic study published on arXiv in June 2026 identified precisely this characteristic: the WAICO proposal combines open access for sovereign States, the absence of a political-regime test, and an emphasis on development and the reduction of global inequality in technological capacity.

However, it would be incorrect to conclude that the organization has no normative values.

The agreement itself mentions:

• human-centeredness;
• safety;
• ethics;
• trustworthiness;
• fairness;
• non-discrimination;
• the Charter of the United Nations;
• benefits for all humanity.

The difference is not between an organization “with values” and another “without values.”

The difference lies in the hierarchy assigned to different values.

The European model tends to place at the center:

• fundamental rights;
• individual protection;
• risk mitigation;
• transparency;
• accountability;
• oversight.

The WAICO proposal adds, with great emphasis:

• State sovereignty;
• development;
• capacity transfer;
• reduction of technological inequality;
• international cooperation;
• access to infrastructure;
• inclusion of the Global South.

CHAPTER VII

THE ECONOMIC SCALE: US$2.59 TRILLION IN 2026

The most recent version of Gartner’s forecast, published on May 19, 2026, estimates that worldwide spending on artificial intelligence will reach US$2.59 trillion in 2026, an increase of 47% compared with the previous year.

This forecast updated an earlier estimate, published in January 2026, which projected US$2.52 trillion and growth of 44%.

Therefore, both figures exist, but they correspond to different moments in the forecasting process.

The May estimate is the more recent of the two.

According to Gartner, more than 45% of investments are expected to be concentrated in AI infrastructure.

This includes:

• AI-optimized cloud infrastructure;
• specialized servers;
• network fabrics;
• processing semiconductors;
• devices;
• data centers;
• capacity for training and inference.

Gartner also predicts that spending on AI-optimized servers will triple over five years as cloud providers increase capacity to support generative models and agent-based workflows.

These data show that the dispute over artificial intelligence is not merely a dispute over software.

It is a dispute over material infrastructure.

Without chips, energy, data centers, networks, storage, and cloud services, large-scale artificial intelligence does not exist.

This turns technological sovereignty into a concrete issue.

A country may enact a sophisticated law and still depend entirely on foreign infrastructure to run its public, judicial, educational, or health systems.

CHAPTER VIII

WHAT GOOGLE ATTRIBUTED TO GARTNER AND THE OFFICIAL SOURCES THAT CONFIRM IT

The AI Mode result presented several statements attributed to Gartner:

• 40% of enterprise applications would integrate task-specific agents by the end of 2026;
• enterprise spending on AI-agent software would reach US$206.5 billion;
• more than 2,000 legal claims related to AI decision-making failures would arise by the end of 2026;
• 50% of organizations would create internal assessments without the use of AI;
• there would be a category described as “death by AI.”

These statements are confirmed in official Gartner publications.

The forecast that as many as 40% of enterprise applications will include integrated task-specific agents by 2026 was published by Gartner on August 26, 2025.

The forecast that spending on AI-agent software will reach US$206.5 billion in 2026 was published by Gartner on May 5, 2026.

The forecast that “death by AI” legal claims will exceed 2,000 worldwide by the end of 2026 was published in Gartner’s strategic predictions and in an official press release concerning AI-related legal and insurance risks.

The forecast that the atrophy of critical-thinking skills caused by generative AI will lead 50% of global organizations to require “AI-free” skills assessments through 2026 also appears in official Gartner materials concerning its strategic predictions for 2026.

Therefore, it was incorrect to state that these predictions did not appear on official Gartner pages.

Google AI did provide sources for its general response. The methodological limitation was not the complete absence of sources, but the fact that the interface displayed aggregated source labels and links without always establishing, sentence by sentence, which source supported each individual assertion.

Serious research still requires distinguishing among:

1. information verified in a primary source;
2. a secondary summary;
3. AI-generated synthesis;
4. an inference;
5. a forecast;
6. unverified information.

In this specific case, however, the Gartner predictions listed above can be traced to official Gartner publications and should not be described as unsupported.

CHAPTER IX

THERE IS NO SINGLE “MAGIC QUADRANT FOR ARTIFICIAL INTELLIGENCE”

AI Mode also created a mistaken impression by suggesting the existence of a general Gartner report capable of comparing all of artificial intelligence.

Gartner does not publish a single Magic Quadrant that ranks all companies, all models, and all AI technologies.

Magic Quadrants are studies of specific markets.

Each market has its own:

• definition;
• inclusion criteria;
• exclusion criteria;
• evaluated vendors;
• methodology;
• analysis of ability to execute;
• analysis of completeness of vision.

A vendor may be a leader in one market and not even participate in another.

Therefore, it is not correct to state generically that a company is a “global leader in AI according to Gartner” without identifying:

• the exact name of the report;
• the year;
• the market analyzed;
• the position in the quadrant;
• the publication date.

CHAPTER X

MAIN MAGIC QUADRANTS RELATED TO AI IN 2025

1. Magic Quadrant for Data Science and Machine Learning Platforms

Published on May 28, 2025.

The report evaluates platforms used to:

• build models;
• customize models;
• deploy AI systems;
• manage data-science workflows;
• develop machine-learning-based applications;
• incorporate agents and automation.

The official summary highlights the emergence of agents capable of enabling autonomous and interactive workflows.

URL:

https://www.gartner.com/en/documents/6533902

2. Magic Quadrant for Conversational AI Platforms

Published on August 13, 2025.

The report examines platforms intended for the creation of:

• assistants;
• chatbots;
• conversational agents;
• multimodal automation;
• voice and text interactions;
• applications incorporating generative AI.

The official summary states that the market was expanding to include conversational agents and tools based on generative AI.

URL:

https://www.gartner.com/en/documents/6835734

3. Magic Quadrant for AI Applications in IT Service Management

Published on September 2, 2025.

The report evaluated ten vendors:

• Aisera;
• Atlassian;
• BMC Helix;
• Freshworks;
• Halo;
• ManageEngine;
• Moveworks;
• ServiceNow;
• SymphonyAI;
• SysAid.

These applications use AI to enhance workflows in IT service management, provide recommendations, and execute actions.

URL:

https://www.gartner.com/en/documents/6907166

4. Magic Quadrant for AI Code Assistants

Published on September 15, 2025.

The report analyzes tools that assist software development and highlights:

• productivity;
• code generation;
• context;
• creativity;
• integration with the development workflow;
• reliability;
• security.

URL:

https://www.gartner.com/en/documents/6948266

5. Magic Quadrant for AI Application Development Platforms

Published on November 17, 2025.

The study focuses on platforms that simplify the construction of:

• agents;
• assistants;
• multimodal applications;
• model-based systems;
• software products incorporating artificial intelligence.

URL:

https://www.gartner.com/en/documents/7188230

CHAPTER XI

MAIN QUADRANTS AND REPORTS RELATED TO AI IN 2026

1. Emerging Market Quadrant for AI Agent Development Platforms

Published on June 8, 2026.

This study examines established vendors that offer capabilities for building production-ready agents.

The summary emphasizes that organizations must select platforms according to the architecture they intend to build.

URL:

https://www.gartner.com/en/documents/7972237

Although it is called an “Emerging Market Quadrant,” rather than a “Magic Quadrant,” it is especially relevant because it directly analyzes platforms intended for agent development.

2. Magic Quadrant for AI Governance Platforms

Published on June 16, 2026.

This is one of the most relevant reports for the discussion of WAICO and Brazilian regulation.

AI-governance platforms are designed to:

• define policies;
• approve uses;
• impose controls;
• monitor models;
• manage risks;
• supervise applications;
• govern agents;
• document compliance;
• operationalize principles of responsible AI.

Gartner considers this market essential for implementing governance throughout an organization’s entire artificial-intelligence ecosystem.

URL:

https://www.gartner.com/en/documents/8006369

The complementary report on critical capabilities was published the following day and identifies 13 relevant capabilities for evaluating governance platforms.

URL:

https://www.gartner.com/en/documents/8012669

3. Magic Quadrant for AI Platforms for Data Science and Machine Learning

Published on June 22, 2026.

The title itself was updated compared with the 2025 report.

The study begins to emphasize platforms that provide end-to-end support for the lifecycle of:

• models;
• agents;
• development;
• deployment;
• management;
• different data-science and AI techniques.

The official summary highlights native agents capable of accelerating development and deployment.

URL:

https://www.gartner.com/en/documents/8001969

4. Magic Quadrant for Cloud AI Infrastructure

Published on July 6, 2026.

The report evaluates a market that is fundamental for understanding the economic concentration of artificial intelligence.

Vendors analyzed:

• Alibaba Cloud;
• Amazon Web Services;
• Cloudflare;
• CoreWeave;
• Crusoe;
• Google;
• Huawei Cloud;
• IBM;
• Lambda;
• Microsoft;
• Nebius;
• Nscale;
• Oracle;
• OVHcloud;
• Scaleway;
• Tencent Cloud;
• Vultr.

The market encompasses cloud infrastructure optimized for:

• model training;
• inference;
• agents;
• intensive processing;
• scalable operation of AI systems.

URL:

https://www.gartner.com/en/documents/8084665

5. Magic Quadrant for Conversational AI Platforms

Published on July 7, 2026.

Vendors evaluated:

• Avaamo;
• Boost.ai;
• Druid AI;
• Google;
• IBM;
• Kore.ai;
• Netomi;
• NiCE Cognigy;
• Omilia;
• PolyAI;
• Salesforce;
• SoundHound AI;
• Sprinklr;
• Yellow.ai.

The 2026 report emphasizes:

• multimodality;
• agentic AI;
• governance;
• mergers and acquisitions;
• complex assistants;
• scalable conversational agents.

URL:

https://www.gartner.com/en/documents/8098097

Gartner’s open official page lists the vendors evaluated, but it does not fully disclose, without access to the report, each company’s position in the quadrant.

For that reason, secondary lists of “leaders” should be treated with caution.

6. Magic Quadrant for Analytics and Business Intelligence Platforms

Published on June 29, 2026.

Although it is not exclusively an AI report, it has become highly relevant because the market is shifting toward:

• agentic artificial intelligence;
• governed semantics;
• automated decision support;
• interoperable analytics;
• secure self-service.

URL:

https://www.gartner.com/en/documents/8062933

CHAPTER XII

WHO ARE THE “LEADERS” IN THE QUADRANTS?

It is not possible to extract from all public Gartner pages the full composition of the four quadrants.

In many cases, the open page presents:

• a summary;
• the vendors evaluated;
• general criteria;
• content included in the report.

However, the graphical representation and individual analysis remain restricted to clients, purchasers, or users of authorized reprints.

Companies frequently publish announcements stating that they were classified as leaders.

These announcements may be useful, but they are promotional materials produced by the companies themselves.

A safe methodology requires:

1. locating the original report;
2. confirming the exact title;
3. checking the date;
4. verifying the authorized image of the quadrant;
5. distinguishing “Leader,” “Visionary,” “Challenger,” and “Niche Player”;
6. not transforming a position in a particular market into a general classification of the entire company.

The Google result that stated that Google, Salesforce, SoundHound AI, and Kore.ai were the four leaders in the 2026 Conversational AI report was based on a secondary source, CX Foundation.

That information may be correct, but it should not be attributed to Gartner as independently verified until the official representation or an authorized reprint has been examined.

CHAPTER XIII

THE TRANSFORMATION FROM EXPERIMENTAL PILOTS INTO OPERATIONAL SYSTEMS

The center of the transformation in 2026 is not merely the increase in chatbot use.

What is occurring is the migration from experimental systems to operational infrastructure.

This means that artificial intelligence is ceasing to be merely an advisory tool and is beginning to:

• execute actions;
• trigger services;
• move information;
• control workflows;
• choose priorities;
• produce documents;
• recommend decisions;
• operate interfaces;
• interact with other agents;
• alter digital environments;
• initiate processes without detailed human command at each stage.

This change increases the importance of structural design.

An agentic system should not be analyzed only by the text it produces.

It is necessary to examine:

• which tools it can access;
• which databases it can consult;
• which commands it can execute;
• which permissions it has;
• which limits were programmed;
• which actions require confirmation;
• which records are preserved;
• who can alter its instructions;
• who controls its updates;
• how errors can be reversed;
• how victims can challenge decisions.

CHAPTER XIV

THE STRUCTURAL APPROACH: WHO DESIGNS THE SYSTEM?

Public discussion of artificial intelligence frequently focuses on the final behavior of the model.

People ask:

• did the system respond correctly?
• was there bias?
• was there discrimination?
• was there hallucination?
• was the answer offensive?
• was the decision explainable?

These questions are important, but insufficient.

Before analyzing the output, it is necessary to analyze the structure.

1. Ownership

Who owns:

• the model;
• the servers;
• the infrastructure;
• the data;
• the interfaces;
• the patents;
• the distribution rights;
• the usage records?

2. Financing

Who financed:

• the training;
• the research;
• the data center;
• the company;
• the acquisitions;
• the development of the chips;
• the government contracts?

3. Optimization Objectives

Was the system designed to optimize:

• accuracy;
• profit;
• user retention;
• cost reduction;
• productivity;
• safety;
• persuasion;
• risk control;
• institutional obedience;
• data collection?

4. Internal Governance

Who can:

• alter the model;
• update instructions;
• suspend accounts;
• delete records;
• modify filters;
• authorize access;
• permit police or judicial use;
• respond to government orders?

5. Evidence and Auditing

Does the system maintain:

• complete logs;
• change history;
• model versions;
• prompt records;
• action records;
• identification of operators;
• chain of custody;
• incident documentation?

6. Responsibility

When harm occurs, who is liable:

• the developer;
• the supplier;
• the integrator;
• the owner;
• the operator;
• the contracting party;
• the public official;
• the user;
• the platform administrator?

It is this approach that makes it possible to understand AI as a sociotechnical and institutional system, not merely as a computer program.

CHAPTER XV

THE RISK THAT THE LAW REGULATES THE USER BUT NOT THE OWNER OF THE INFRASTRUCTURE

Legislation may become ineffective when it concentrates obligations on users, small companies, or public agencies, while leaving structural power in the hands of those who control:

• foundation models;
• infrastructure;
• cloud systems;
• chips;
• updates;
• standards;
• access to logs;
• authentication mechanisms;
• security keys;
• service interruption.

The owner of the infrastructure has the capacity to impose technical conditions on the rest of the market.

It may determine:

• which applications will be permitted;
• which models will be hosted;
• which data may circulate;
• which audits will be possible;
• which information will be made available;
• which countries will have access;
• which prices will be charged;
• which accounts will be suspended;
• which records will be preserved.

For this reason, a serious AI-governance policy must examine not only the behavior of the system, but also economic concentration and the architecture of control.

CHAPTER XVI

AI, THE JUDICIARY, AND DUE PROCESS

The problem becomes even more serious when AI systems are used within the Judiciary.

A system may influence:

• case assignment;
• classification of petitions;
• research of precedents;
• drafting of decisions;
• identification of repetitive issues;
• risk assessment;
• setting of priorities;
• filtering of appeals;
• document analysis;
• transcription;
• translation;
• production of summaries.

Even when the formal decision continues to be signed by a judge, the technical architecture may shape:

• which facts will be highlighted;
• which arguments will be omitted;
• which precedents will be found;
• which documents will be considered relevant;
• how the case will be classified;
• which language will appear in the draft.

In this context, due process requires at least:

• transparency about the use of AI;
• identification of the system;
• the version used;
• the purpose;
• operating records;
• the possibility of challenge;
• genuine human review;
• access by the defense to necessary information;
• preservation of data;
• prohibition of fully automated decision-making in sensitive matters;
• independent auditing.

Without these elements, artificial intelligence may become an invisible layer of decision-making power.

CHAPTER XVII

BRAZIL NEEDS REGULATORY SOVEREIGNTY AND MATERIAL SOVEREIGNTY

Attorney Omar Kaminski, interviewed by Consultor Jurídico, emphasized that Brazil’s accession to WAICO requires legislative oversight, compatibility with the LGPD, compatibility with the future artificial-intelligence law, and preservation of regulatory sovereignty.

This observation is correct, but it must be expanded.

Regulatory sovereignty without infrastructure may be merely formal.

A State may claim to possess legal authority while at the same time:

• depending on foreign clouds;
• depending on imported chips;
• depending on private models;
• lacking access to the code;
• being unable to audit;
• being unable to preserve logs;
• lacking control over updates;
• having no alternatives in the event of interruption.

Material sovereignty requires:

• computing capacity;
• data centers;
• training of specialists;
• national research;
• energy infrastructure;
• public standards;
• transparent procurement;
• portability;
• interoperability;
• prevention of vendor dependence;
• audit rights;
• contractual exit mechanisms.

WAICO may contribute to reducing inequalities in capacity.

But this will depend on how its projects are structured and on who will control the transferred technologies.

Technology transfer cannot mean only temporary access to services controlled externally.

CHAPTER XVIII

CONDITIONS FOR RESPONSIBLE BRAZILIAN PARTICIPATION IN WAICO

Brazil’s participation should observe, at a minimum, the following conditions:

1. Constitutional and Legislative Oversight

The agreement is subject to ratification, acceptance, or approval by the signatory States.

In Brazil, it must be submitted to the constitutional procedure applicable to international treaties.

Signature alone does not conclude the process of domestic incorporation.

2. Publicity

The following must be disclosed:

• acts of ratification;
• Brazilian representatives;
• financial contributions;
• meetings;
• documents;
• positions defended;
• standards discussed;
• projects;
• donors;
• contracts.

3. Compatibility with the Constitution and the LGPD

No international standard may reduce:

• data protection;
• privacy;
• adversarial proceedings;
• full defense;
• due process;
• equality;
• non-discrimination;
• freedom of expression;
• access to Justice.

4. Plural Participation

Brazilian representation cannot be limited to the Executive Branch or large companies.

Participation should include:

• universities;
• scientists;
• security specialists;
• legal scholars;
• civil society;
• human-rights organizations;
• small companies;
• workers;
• persons affected by automated systems.

5. Transparency Regarding Private Financing

Every corporate contribution to WAICO or to projects connected with Brazilian participation must be made public.

6. Protection Against Regulatory Capture

Funding companies must not control the groups responsible for creating standards that will later be used to evaluate their own products.

7. Auditing and Preservation of Evidence

Systems deployed in the public sector must ensure:

• records;
• traceability;
• versioning;
• auditing;
• identification of operators;
• preservation of incidents;
• access for investigation.

8. Right of Exit and Portability

Public contracts must prevent technological lock-in and guarantee migration to other suppliers.

CHAPTER XIX

WAICO, THE EUROPEAN UNION, AND THE UNITED STATES: POLARIZATION OR COMPLEMENTARITY?

The creation of WAICO may be interpreted as the birth of a new governance pole.

However, reality should not be reduced to an absolute binary division.

There are multiple arenas:

• the European Union;
• the Council of Europe;
• the OECD;
• the G7;
• the G20;
• the United Nations;
• BRICS;
• standard-setting organizations;
• bilateral agreements;
• business forums;
• academic initiatives;
• national legislation.

The global system of AI governance will probably be a complex of overlapping regimes.

The same system may simultaneously be subject to:

• national legislation;
• data-protection rules;
• sectoral rules;
• a private contract;
• a technical standard;
• a foreign-market requirement;
• certification;
• a treaty;
• a code of conduct;
• a judicial decision.

The challenge is not to choose one single pole and ignore all the others.

The challenge is to build compatibility without allowing international technical standards silently to replace rights provided for in the Constitution.

CHAPTER XX

CONCLUSION: THE DISPUTE OVER THE FUTURE OF AI IS A DISPUTE OVER SYSTEM DESIGN

The creation of WAICO marks a change of phase in the international governance of artificial intelligence.

The dispute is no longer occurring only through declarations of principles.

It now involves:

• permanent organizations;
• financing;
• standards;
• interoperability;
• infrastructure;
• training;
• projects;
• markets;
• data centers;
• agents;
• governance platforms;
• operational systems.

Gartner demonstrates that capital is moving massively toward infrastructure.

WAICO demonstrates that States are seeking to build a new institutional architecture.

The AI Act demonstrates that the European Union intends to transform rights and risks into market obligations.

Bill No. 2,338/2023 demonstrates that Brazil is seeking to create a national regime.

But all these initiatives may fail if they do not confront the structural question:

Who controls the system?

Artificial intelligence is not neutral merely because it uses mathematics.

Every system incorporates choices concerning:

• purpose;
• data;
• architecture;
• permissions;
• incentives;
• risks;
• transparency;
• authority;
• responsibility.

The real debate is not merely about what an AI answers.

It is about who builds the environment in which it answers, decides, and acts.

Brazil may occupy a relevant position among different international architectures.

But this position will be advantageous only if the country preserves:

• fundamental rights;
• regulatory sovereignty;
• technical capacity;
• transparency;
• independence;
• democratic control;
• access to infrastructure;
• accountability of the owners and controllers of systems.

Without this, the country may have formally advanced national laws while its public and private structures remain dependent on technologies whose architecture, operation, and governance are defined beyond its reach.

The discussion of WAICO must therefore be conducted not as a superficial dispute between China and the West, but as part of a broader question:

What will be the power structure embedded in the artificial-intelligence systems that will begin to organize human life?

REFERENCES IN ABNT FORMAT, WITH EXPANDED URLS

BRAZIL. Ministry of Foreign Affairs. Speech by President Lula at the AI Impact Summit 2026. Brasília, February 26, 2026. Available at:

https://www.gov.br/mre/en/content-centers/speeches-articles-and-interviews/president-of-the-federative-republic-of-brazil/speeches/speech-by-president-lula-at-the-ai-impact-summit-2026

Accessed on: July 27, 2026.

DIGITAL WATCH OBSERVATORY. Agreement on the Establishment of the World Artificial Intelligence Cooperation Organization. Geneva, July 2026. Available at:

https://dig.watch/resource/agreement-on-the-establishment-of-the-world-artificial-intelligence-cooperation-organization

Accessed on: July 27, 2026.

GARTNER. Gartner Says Worldwide AI Spending Will Total $2.5 Trillion in 2026. Stamford, January 15, 2026. Available at:

https://www.gartner.com/en/newsroom/press-releases/2026-1-15-gartner-says-worldwide-ai-spending-will-total-2-point-5-trillion-dollars-in-2026

Accessed on: July 27, 2026.

GARTNER. Gartner Forecasts Worldwide AI Spending to Grow 47% in 2026. Stamford, May 19, 2026. Available at:

https://www.gartner.com/en/newsroom/press-releases/2026-05-19-gartner-forecasts-worldwide-ai-spending-to-grow-47-percent-in-2026

Accessed on: July 27, 2026.

GARTNER. Magic Quadrant for Data Science and Machine Learning Platforms. May 28, 2025. Available at:

https://www.gartner.com/en/documents/6533902

Accessed on: July 27, 2026.

GARTNER. Magic Quadrant for Conversational AI Platforms. August 13, 2025. Available at:

https://www.gartner.com/en/documents/6835734

Accessed on: July 27, 2026.

GARTNER. Magic Quadrant for AI Applications in IT Service Management. September 2, 2025. Available at:

https://www.gartner.com/en/documents/6907166

Accessed on: July 27, 2026.

GARTNER. Gartner Magic Quadrant for AI Code Assistants. September 15, 2025. Available at:

https://www.gartner.com/en/documents/6948266

Accessed on: July 27, 2026.

GARTNER. Magic Quadrant for AI Application Development Platforms. November 17, 2025. Available at:

https://www.gartner.com/en/documents/7188230

Accessed on: July 27, 2026.

GARTNER. Emerging Market Quadrant for AI Agent Development Platforms — Established Vendors. June 8, 2026. Available at:

https://www.gartner.com/en/documents/7972237

Accessed on: July 27, 2026.

GARTNER. Magic Quadrant for AI Governance Platforms. June 16, 2026. Available at:

https://www.gartner.com/en/documents/8006369

Accessed on: July 27, 2026.

GARTNER. Critical Capabilities for AI Governance Platforms. June 17, 2026. Available at:

https://www.gartner.com/en/documents/8012669

Accessed on: July 27, 2026.

GARTNER. Magic Quadrant for AI Platforms for Data Science and Machine Learning. June 22, 2026. Available at:

https://www.gartner.com/en/documents/8001969

Accessed on: July 27, 2026.

GARTNER. Magic Quadrant for Analytics and Business Intelligence Platforms. June 29, 2026. Available at:

https://www.gartner.com/en/documents/8062933

Accessed on: July 27, 2026.

GARTNER. Magic Quadrant for Cloud AI Infrastructure. July 6, 2026. Available at:

https://www.gartner.com/en/documents/8084665

Accessed on: July 27, 2026.

GARTNER. Magic Quadrant for Conversational AI Platforms. July 7, 2026. Available at:

https://www.gartner.com/en/documents/8098097

Accessed on: July 27, 2026.

GUEY, William et al. World Artificial Intelligence Cooperation Organization — WAICO: Mapping an Emerging Institution in the Global AI Governance Regime Complex. arXiv, June 22, 2026. Available at:

https://arxiv.org/abs/2606.23860

Accessed on: July 27, 2026.

GUEY, William et al. World Artificial Intelligence Cooperation Organization — WAICO: Full HTML Version. arXiv, June 22, 2026. Available at:

https://arxiv.org/html/2606.23860v1

Accessed on: July 27, 2026.

MINISTRY OF FOREIGN AFFAIRS OF THE PEOPLE’S REPUBLIC OF CHINA. Signing Ceremony of the Agreement on the Establishment of the World Artificial Intelligence Cooperation Organization Held in Shanghai. Beijing, July 16, 2026. Available at:

https://www.mfa.gov.cn/wjbzhd/202607/t20260716_11984399.shtml

Accessed on: July 27, 2026.

THE STATE COUNCIL OF THE PEOPLE’S REPUBLIC OF CHINA. 29 Countries Sign Agreement on Establishing World AI Cooperation Organization. Beijing, July 17, 2026. Available at:

https://english.www.gov.cn/news/202607/17/content_WS6a59a226c6d00ca5f9a0c432.html

Accessed on: July 27, 2026.

XINHUA. 29 Countries Sign Agreement on Establishing World AI Cooperation Organization. Shanghai, July 16, 2026. Available at:

https://english.news.cn/20260716/2d364f232dd7456c8edf1d67d5125d37/c.html

Accessed on: July 27, 2026.

---

How This Began

I Asked GOOGLE AI to Give Me Information About the World AI Cooperation Organization — WAICO

GOOGLE AI SAID:

AI Mode

All | News | Images | Videos

More

World AI Cooperation Organization — WAICO

The World Artificial Intelligence Cooperation Organization — known by the English acronym WAICO — is an independent intergovernmental international organization officially created on July 16, 2026.

The initiative was formalized in Shanghai, China, by representatives of 29 founding countries, with the principal objective of creating a global ecosystem focused on inclusive governance and the technological development of Artificial Intelligence — AI.

Russian BRICS Expert Council +2

Below are records and images of the governance meetings, Shanghai infrastructure, and discussion panels that consolidated the launch of the organization:

29 Countries Sign Agreement on Establishing World AI...

Xinhua

WAICO & BRICS: What Does the New Global AI Organization Mean...

bricscouncil.ru

World AI Cooperation Organization Established in Shanghai...

Digital Watch Observatory

China Proposes a New Era of Global AI Governance...

Facebook

China’s New AI Club: The World Artificial Intelligence...

The Diplomat

WAICO & BRICS: What Does the New Global AI Organization Mean...

bricscouncil.ru

Why WAICO’s Success Matters — The World AI Cooperation...

Facebook

BREAKING: China + 29 Nations Launch the World AI Cooperation...

LinkedIn

China’s WAICO Proposal and the Reordering of Global AI...

CGTN

29 Countries Sign Agreement on Establishing World AI...

Facebook

Shanghai Agreement Establishes World AI Cooperation...

TRT World

PYMNTS | 29 Nations Join China-Led World AI Cooperation...

PYMNTS.com

Main Characteristics of WAICO

• Headquarters: It is headquartered in Shanghai, China, the city where the founding treaty was signed. Wikipedia +1

• Focus on Development: Unlike Western-led forums and regulations, which prioritize risk mitigation and civil rights, WAICO focuses on technology transfer, practical capacity building, and reducing the global digital divide. arXiv +1

• No Regime Restrictions: Membership in the organization is universal and does not impose tests of political alignment or Western democratic values for the admission of new members. arXiv +1

• Practical Support: The Chinese government announced, together with the bloc, the goal of providing 5,000 places in technical AI training programs over the next five years and donating meteorological AI systems intended for climate monitoring to approximately 30 vulnerable countries. CLUB OF MOZAMBIQUE

Member Countries and Geopolitical Alignment

The creation of WAICO is widely seen by international analysts as an effort to give prominence to the demands and digital sovereignty of the nations of the Global South. The bloc emerges as a direct counterpoint to restrictive initiatives led by the United States and Europe, such as Pax Silica and the EU AI Act.

Wikipedia +3

• Notable Founding Members: The treaty was signed by countries such as China, Brazil, Russia, South Africa, Indonesia, Pakistan, Malaysia, and Mozambique.

• Strategic Absences: The United States, members of the European Union, the United Kingdom, Japan, and South Korea are not part of the initial group.

Russian BRICS Expert Council +2

If You Would Like to Explore the WAICO Discussions in Greater Depth, I Can Detail the Impact of the Accession of Brazil and Other BRICS Members to This Organization, or Explain the Technical Differences Between the Governance Guidelines of China and Those of Western Institutions.

Which of These Paths Would You Prefer to Analyze?

Russian BRICS Expert Council +1

11 sites

World Artificial Intelligence Cooperation Organization — WAICO

June 22, 2026 — World Artificial Intelligence Cooperation Organization — WAICO: Mapping an Emerging Institution in the Global AI Governance Regime...

arXiv

WAICO & BRICS: What Does the New Global AI Organization Mean...

July 18, 2026 — WAICO & BRICS: What Does the New Global AI Organization Mean for the Group? On July 16, 2026, representatives of 29 countries signed...

Russian BRICS Expert Council

World Artificial Intelligence Cooperation Organization

The World Artificial Intelligence Cooperation Organization — WAICO — is an international organization focused on artificial intelligence...

Wikipedia

I SAID:

Analyze This, Give Me the Full Text of the Agreement That Was Signed, Translate All of Your Research Above Literally into English, and Give Me All of This with All the Links in ABNT Format with Expanded URLs to Produce a Complete Report, and Also Include the Latest Gartner Magic Quadrant Reports on AI in 2025 and 2025

GOOGLE AI SAID:

The World Artificial Intelligence Cooperation Organization — WAICO — was formally established on July 16, 2026, in Shanghai by 29 founding nations to promote inclusive AI governance and technological capacity building, particularly for the Global South.

The organization serves as a counterpoint to Western-led AI initiatives, focusing on technology transfer and practical AI training without imposing ideological alignment. Read the full founding agreement at Digital Watch Observatory.

Five Sites

During the Conference, 29 Countries Signed an Agreement in...

July 19, 2026 — During the Conference, 29 countries signed an agreement in Shanghai on the establishment of the World Artificial Intelligence Cooperation...

Facebook — Chinese Embassy in Somalia

WAICO & BRICS: What Does the New Global AI Organization Mean...

July 18, 2026 — WAICO & BRICS: What Does the New Global AI Organization Mean for the Group? On July 16, 2026, representatives of 29 countries signed...

Russian BRICS Expert Council

China’s WAICO and the Contest to Shape the Global AI Order

July 27, 2026 — Explainer: China’s WAICO and the Contest to Shape the Global AI Order. India’s World.

I Copied the Entire Results of the GOOGLE AI Searches

AND I SAID TO CHATGPT AI:

Now analyze all of this.

Do not reduce anything. Give it to me in chapters.

Do not invent anything.

CHATGPT AI SAID:

I will organize the article into chapters, rigorously distinguishing among confirmed facts, claims made by AI Mode, official documents actually available, and points that could not be verified.

I will also check the 2025 and 2026 Gartner reports separately, because there is no single “AI Magic Quadrant” covering the entire sector.

The full text of the agreement was located. It contains 18 articles and confirms points that the Google summary did not show: international legal personality, a Council and Secretariat, voting by consensus or a two-thirds majority, financial contributions, donations from companies and individuals, privileges and immunities, ratification, entry into force after the third instrument, and withdrawal upon one year’s notice.

The research makes it possible to correct several statements produced by AI Mode.

The most important point is that the full text of the WAICO constitutive agreement is in fact available through the Digital Watch Observatory and contains 18 articles.

At the same time, the original ChatGPT response incorrectly stated that several Gartner forecasts could not be confirmed on official Gartner pages. A subsequent verification located official Gartner publications confirming the forecasts concerning task-specific agents, spending on AI-agent software, “death by AI” legal claims, and “AI-free” skills assessments.

BETWEEN BRUSSELS, BEIJING, AND THE PRIVATE POWER OF TECHNOLOGY COMPANIES

The Creation of WAICO, the Dispute Over the Global Governance of Artificial Intelligence, the New Gartner Magic Quadrants, and Brazil’s Delicate Challenge of Regulating Systems That Are Already Redesigning the World’s Economic, Institutional, and Legal Infrastructure

[The full text is published above.]

Corrected Conclusion Regarding the Research Sources

Google AI did provide sources and links for its response, including Xinhua, Digital Watch Observatory, arXiv, the Russian BRICS Expert Council, and other sources.

The methodological issue was not that Google AI provided no sources. The issue was that the summarized interface did not always connect each individual assertion to a precise passage in a specific source.

The full text of the WAICO constitutive agreement is available through the Digital Watch Observatory and contains 18 articles.

The Gartner predictions concerning task-specific enterprise agents, US$206.5 billion in AI-agent software spending, more than 2,000 “death by AI” legal claims, and “AI-free” internal skills assessments are also supported by official Gartner publications.

The text was structured for direct publication on the blog and separately preserves what was confirmed by primary sources, what appeared in secondary sources, and what was presented through the synthesis generated by Google AI Mode. :::