New technologies, especially artificial intelligence (AI), are transforming information ecosystems worldwide, reshaping how information is produced, accessed and shared. However, these rapid technological developments raise questions about how we can ensure that the information we rely on is trustworthy. Ultimately, AI’s impact will depend on the choices made by governments, technology companies, and society.
The United Nations Secretary-General, António Guterres, has urged that AI “cannot be built on hype – or disinformation. We need facts we can trust – and share – across countries and across sectors. Less noise. More knowledge.”
The development of AI technologies is outpacing existing governance frameworks and raising fundamental questions about accountability, equality, safety and security. To address this gap, the United Nations is working to strengthen global AI governance and place science at the centre of international cooperation on AI.
What is information integrity?
As outlined in the UN report, From Principles to Practice: Strengthening Information Integrity, “information integrity refers to an information ecosystem in which reliable and accurate information is available to all, enabling people to engage meaningfully in public life, make informed decisions and exercise their rights. This ecosystem is shaped by the actions of a diverse range of actors, including governments, technology companies, media, civil society and individuals.”
These information spaces are increasingly under strain because of information manipulation.
How can the use of AI threaten information integrity?
Generative AI tools are transforming how people create and consume content, from editing text to generating images and videos. AI systems are increasingly integrated into digital platforms. While these tools can enhance efficiency and creativity, they also pose risks to information integrity and human rights.
Threats to information integrity include misinformation (the unintentional sharing of inaccurate information), disinformation (information deliberately created or disseminated to cause harm) and hate speech (the use of discriminatory language referring to someone’s religion, ethnicity, nationality, race, colour, descent, gender or other identity factor).
Information risks undermine the integrity of information ecosystems, potentially reducing public access to evidence-based information and weakening trust and social cohesion.
The main risks to information integrity linked to AI also include:
- AI’s role in curating and generating information, and growing reliance on these systems to shape people’s understanding of the world. As generative AI tools cannot be relied on as sources of accurate information, their use can contribute to declining trust in information sources more broadly.
- The misuse of AI to create and spread false or hateful content at an unprecedented scale. Generative AI tools have made it much easier (and less costly) to spread false claims for financial or strategic gain. These hateful messages often target vulnerable groups or individuals.
Generative AI technology can produce content that appears highly credible while containing false or misleading information, contributing to trust deficits. AI systems generate outputs based on patterns learned from their training data. These outputs can reflect both accurate and inaccurate information, making it harder to distinguish reliable information from false or misleading content.
The growing use of synthetic media, often referred to as “deepfakes”, has also made it more difficult to distinguish authentic content from fabricated material without technological assistance. This can further undermine trust in digital information ecosystems.
Algorithms used by digital platforms may prioritise engagement over accuracy, potentially amplifying false or misleading content. Although fact-checking technologies are becoming more sophisticated, there are still no universal systems to verify content credibility, raising concerns about authenticity, bias, and information privacy. Together, these threats can weaken human rights protections and undermine democratic institutions.
A lack of diversity in training data can cause AI systems to generate misleading information and reinforce existing inequalities.
In April 2026, the UN also warned that unchecked AI use in advertising could accelerate the global information-integrity crisis, as advertising revenue may continue to support online content regardless of its accuracy or reliability. The UN called for stronger safeguards and highlighted the role advertisers and major brands can play in improving the information ecosystem.
Finally, AI can also affect information integrity through unequal access to technology and decision-making. As the UN Secretary-General warned at the opening of the first Global Dialogue on AI Governance: “When power imbalances are hard-wired into technology, inequality becomes part of the code.” Access to computing power, data and technical expertise remains concentrated among a relatively small number of companies and countries. Ensuring that AI governance is inclusive and that its benefits are widely shared is therefore essential to building healthy information ecosystems.
How can we mitigate these risks?
AI-related risks can be reduced by prioritising transparency and fairness throughout the life cycle of AI systems, from design and development to deployment and decommissioning. AI systems should be trained on reliable and inclusive information sources. Independent audits of AI models by researchers should be encouraged. Data on AI model risks should be publicly accessible, particularly information related to discrimination or so-called ‘hallucinations’ (content that appears factual but is fabricated).
The development and deployment of AI systems should also be guided by accountability, while AI algorithms should operate within ethical frameworks. AI-generated content should be assessed for factual accuracy and bias before dissemination, and digital platforms that use AI to curate and promote content should be transparent about how these systems operate and be held accountable for their impacts.
Clear policies on content provenance are also needed, including both visible and invisible forms, such as authenticity certification, watermarking and labelling. Strengthening AI literacy and public understanding of how AI systems function can also help mitigate risks.
AI tools could also support content monitoring to help counter hate speech and disinformation, and assist content moderators. Producing content in local languages and multilingual content moderation are particularly important, especially in conflict-affected and fragile settings, as they can improve access to reliable information.
What does the UN do to safeguard information integrity?
The UN has emphasized the importance of ethical guidelines and strengthened governance to ensure that AI supports, rather than undermines, information integrity.
The 2023 policy brief, Information Integrity on Digital Platforms, highlights how threats to information integrity are impacting progress on global, national and local issues. It outlines principles for a code of conduct to help make the digital space more inclusive and safer for all.
In June 2024, the UN launched the “Global Principles for Information Integrity”, emphasizing the need for immediate action to address the harms caused by mis- and disinformation, as well as hate speech. This framework aims to galvanize international cooperation and foster safer and more inclusive digital environments. The UN has called on governments, technology companies, media and civil society to promote transparency, independent media, and human rights.
In March 2024, the General Assembly adopted its first resolution on the promotion of “safe, secure and trustworthy” AI systems that respect human rights and comply with international law.
In September 2024, the Global Digital Compact (GDC) was adopted as part of the Pact for the Future. It aims to build an inclusive, open, and safe digital space and to promote independent, fact-based, and timely information to counter mis- and disinformation. It also includes commitments by UN Member States to close the digital divide, promote universal digital public infrastructure, and strengthen AI governance.
A key outcome of the Global Digital Compact is the establishment of an Independent International Scientific Panel on AI. The Panel provides independent, multidisciplinary, evidence-based scientific assessments to support international cooperation on AI governance and improve understanding of AI’s opportunities, risks and impacts.
In July 2026, the Panel published its first preliminary report, which concludes that current safeguards cannot keep pace with the growth of AI’s capabilities.
The Compact also established a Global Dialogue on AI Governance. In July 2026, the first Global Dialogue on AI Governance brought governments and stakeholders together to discuss how AI development can benefit everyone.
Protecting information integrity is a shared responsibility of governments, technology companies, media, civil society and individuals. The United Nations is committed to ensuring that AI becomes a force for good, supporting democracy, human rights and access to reliable information.
Further reading
Artificial Intelligence (A.I.)
Artificial Intelligence – Selected Online Resources
UNRIC Info Point & Library Newsletter – October 2025 (TOPIC OF THE MONTH: Information Integrity)
Artificial Intelligence – AI | UNESCO