A number of international organizations have released materials concerning the use of information technologies for anti-corruption purposes:

1) APEC presented the report “Technologies for Preventing, Detecting, and Combating Corruption”, which examines the mechanisms for the use of IT already implemented in member economies (for example, e-government, online public procurement portals, open databases and registries, whistleblower platforms) and possible promising directions for their application (including the use of artificial intelligence (AI) / machine learning (ML) to detect anomalies and predict corruption risks; implementation of blockchain registries to record immutable traces of decisions and contracts; the use of advanced data analytics and network analysis for investigations; the use of facial recognition and biometrics for access control; the use of drones and remote monitoring to ensure compliance in hard-to-reach areas; integration of several technologies to build autonomous systems for preventing violations).

2) OECD released the report “Governing with Artificial Intelligence: The State of Play and Way Forward in Core Government Functions”, which examines, among other things, existing examples of the use of IT and AI for anti-corruption purposes, possible risks of their application, and key areas for their further implementation.
In particular, the authors divide current directions of use into three key groups:

a) detecting fraudulent and corrupt activity (examples include a pilot data analysis and machine-learning system in Spain to identify anomalies associated with corruption and fraud risks; the use of ML to analyze the EU Transparency Register and identify suspicious registrations at the European Court of Auditors);

b) improving efficiency in knowledge management and information synthesis (for example, in Brazil, the introduction of the AI assistant ChatTCU, based on GPT-4, into the work of the Federal Court of Accounts has allowed auditors to quickly obtain and synthesize information, reducing manual workload while maintaining accuracy and security);

c) enhancing predictive analytics and risk modeling (for example, in Bogotá (Colombia), the VigIA system analyzes contracts of the mayor’s office and predicts the probability of corruption; in Italy, algorithms predict economic crimes based on police archives; in Lithuania, work is underway on a tool that uses large language models (LLMs), specifically trained on data on corruption risks and risk-assessment methodologies, to help anti-corruption officials quickly and effectively analyze regulations and draft laws in order to detect potential shortcomings or vulnerabilities from the standpoint of corruption risks).

Among the challenges and risks of using IT and AI in anti-corruption activities, the report notes insufficient or distorted data (errors in initial data may lead to bias and incorrect conclusions), lack of transparency and explainability of algorithms (“black-box” functioning of machine-learning models undermines public trust and raises concerns among auditors), erroneous or unethical use of AI, such as the processing of sensitive personal data without adequate protection.

Promising areas for development include:

• developing adaptive learning models that update risk analysis in real time;
• using analysis of non-numerical data and sensor technologies (combining AI with satellite, drone, thermal-sensor, and radar data), which would make it possible to track corruption schemes and violations (for example, land-use or construction oversight);
• applying agent-based modeling (ABM), which simulates the behavior of public officials, companies, and other actors to predict the behavior of corruption networks;
• AI-based analysis of influence networks, identifying hidden connections among politicians, businesses, and lobbyists.

In addition, the OECD Business and Industry Advisory Committee (BIAC) presented the report “Harnessing AI for Integrity: Opportunities, Challenges, and the Business Case Against Corruption”. The authors examine how artificial intelligence can transform anti-corruption efforts and strengthen integrity systems in the public and private sectors.
Examples highlighted include continuous-monitoring systems that analyze 100% of transactions and procurements (for example, the RITA tool in Italy); algorithms for detecting cartels and anti-competitive collusion (such as BRIAS in South Korea); digital public-administration and service-delivery platforms (Estonia); AI-supported due-diligence tools for clients and counterparties in large corporations, etc.

At the same time, the report notes that businesses – especially small and medium-sized enterprises – face several barriers when adopting AI, including: a complex regulatory environment and cross-country differences; insufficient digital infrastructure; risks of data bias, algorithmic errors, and shortage of qualified personnel; low readiness of corporate processes for AI integration; security and data-protection issues.

And the main theme of this year’s OECD Global Anti-Corruption and Integrity Forum was the use of digital technologies to enhance transparency, detect conflicts of interest, identify corruption risks, and support other areas related to anti-corruption efforts.

3) At the UN Second Global Conference on Harnessing Data to Improve Corruption Measurement, one of the sessions discussed how AI and ML can prevent and address corruption, while examining ethical, technical and operational challenges, especially in contexts with limited data or weak digital infrastructure, and reviewing current practices, capacity needs, governance safeguards and opportunities for greater collaboration in leveraging AI for corruption measurement.

And the UN Office on Drugs and Crime (UNODC) this year will dedicate its Academic symposium, to be held in Qatar on 13–14 December 2025, to the topic of AI use in anti-corruption efforts. The event will include several thematic tracks, both research-oriented and applied:

• AI in the Public Sector: Strengthening Integrity Systems from the Top Down;
• From the Ground Up: Bottom-Up Uses of AI in the Fight Against Corruption;
• Mapping Corruption: AI, Satellite Data, and the Geography of Accountability;
• Lightning Talks: Frontiers of Interdisciplinary Anti-Corruption Research;
• Interdisciplinary Poster Session by Emerging Researchers;
• From Code to Impact: Applied AI Projects in Anti-Corruption Practice.

4) The International Anti-Corruption Academy (IACA) held the webinar “Is AI Rewriting the Fight Against Corruption? From Predictive Prevention to Cross-Border Forensics – Rethinking Strategy with AI”, which was joined by more than 120 participants. During the event, experts discussed how large language models and other AI tools are used by governments to enhance transparency, strengthen financial integrity, and detect corruption, as well as the advantages and risks of AI implementation.

NGO’s have also joined the study of AI applications for combating corruption:

1) The U4 Anti-Corruption Resource Centre presented a brief analytical note “Artificial Intelligence in Anti-Corruption – A Timely Update on AI Technology”, which we have already covered on our Portal, as well as the review developed jointly with Transparency International (TI): “Harnessing Artificial Intelligence (AI) for Anti-Corruption”, which summarizes key benefits and challenges of using AI to prevent, detect, and investigate corruption.

In this review, the authors:

• provide a brief overview of various AI technologies, including ML, deep learning, natural-language processing (NLP), computer vision (CV), and generative AI;
• examine key advantages and challenges of using AI in anti-corruption. General advantages include: a) autonomous learning and scalability, b) use of advanced computing power, c) reduced need for human intervention, d) ability to integrate new and fragmented data sources; main challenges include: a) unreliable data and bias, b) algorithmic problems and potential manipulation, c) the need for human oversight, d) institutional unpreparedness and regulatory lag, e) gaps in inclusivity, f) information asymmetry and opacity;
• describe broader issues and implications of using AI in anti-corruption, including:

1. increased surveillance and risks to rights and freedoms – the use of AI may lead to excessive monitoring of citizens and public officials, creating privacy risks and enabling politically motivated control;
2. increased existing inequalities – if data contain historical biases, AI reproduces and amplifies them, leading to discrimination and a higher likelihood of unjust inclusion of vulnerable groups in “risk categories”;
3. institutional dependence on AI – excessive reliance on automated outputs reduces the role of professional judgment, weakens critical thinking, and may lead to errors that go unnoticed due to the perceived authority of technology;
4. opacity and lack of control – black-box algorithms make decisions difficult to explain, reducing the ability of citizens and NGOs to exercise oversight over public institutions;
5. undermining legitimacy and trust – algorithmic errors or bias undermine trust in anti-corruption bodies and may lead to challenges to decisions, including judicial decisions;
6. corrupt use of AI – public officials or other interested actors may use AI for personal gain: to hide traces of wrongdoing, target opponents, or manipulate data;
7. technical vulnerability to attacks and manipulation – AI models may be “poisoned” with malicious data, hacked, or otherwise manipulated externally;
8. increasing costs and technological inequality – high infrastructure, staffing, and resource requirements widen the gap between organizations and countries able to use AI and those left behind.

2) Transparency International also released a thematic brief “Addressing Corrupt Uses of Artificial Intelligence”, which analyzes the potential malicious use of AI by public officials.
The authors note that AI is vulnerable to corruption due to several features, including opacity of algorithms (“black box”), dependence on hidden training data, illusion of objectivity, concentration of power among those who control algorithms, deployment through non-transparent procurement, and the extensive capacity to automate corrupt schemes at scale.

Among the most common forms of corrupt use of AI, they highlight:

• manipulation of automated decision-making – deliberate distortion of data, algorithm parameters, or interpretation of outputs;
• using AI to conceal or launder corruption proceeds – generating fake documents, identities, and complex transaction schemes;
• manipulation of public opinion – creating disinformation, political deepfakes, and targeted influence campaigns;
• suppression of civil society and opposition – using AI for surveillance, identifying activists, manipulating content visibility, and limiting public oversight.

TI also offers recommendations for the use of AI:

• for governments – develop national AI strategies with an anti-corruption section, strengthen transparency (maintain algorithm registries, increase openness of AI-technology procurement), ensure explainability and oversight of AI systems, protect elections from abuse, introduce mechanisms for challenging AI-made decisions and provide independent oversight;
• for companies and AI providers – conduct due diligence on public-sector clients, monitor and prevent corrupt use of their systems, enable auditing, report abuses, and cooperate with authorities and the public;
• for civil society – build AI expertise, participate in drafting AI-related legislation, monitor government AI projects, conduct public-awareness activities, and assess regulations for risks of corrupt use.

In addition, Transparency International this year held the forum “AI and Data Integrity for Compliance”, dedicated to risks, opportunities, and practical applications of AI in the business sector.

At the same time, academic and research publications on the application of IT and AI for anti-corruption purposes are growing rapidly – in scholarly periodicals and at thematic forums. In this regard, our Portal has a separate subsection in the Literature section, which brings together books and articles devoted to the use of digital technologies in the field of combating corruption.