The study Anticipating Future Corruption: Strengthening Anti-Corruption through Foresight and Anticipatory Governance is grounded in the understanding of corruption as a complex and constantly evolving system, rather than as a collection of isolated violations that can be addressed through specific targeted measures. The authors argue that corrupt practices are shaped by political, economic, social, legal and technological factors and sustained through networks of interconnected actors. These networks are themselves able to adapt to changes in the legal framework, new technologies and stronger enforcement by changing their membership, modes of interaction and corruption strategies.

According to the authors, the traditional linear approach, which assumes that a particular anti-corruption measure will produce a predetermined result, does not always account for this capacity to adapt. As a result, anti-corruption measures may not only prove ineffective but may also generate unintended consequences – for example, by displacing corruption risks elsewhere or prompting the emergence of new ways to circumvent controls.

One of the examples examined in the report is corruption at the Port of Rotterdam, where stronger enforcement against drug trafficking, together with the digitalisation and automation of controls, did not eliminate corruption risks but changed their nature: bribery of inspectors increasingly gave way to collusion and coercion targeting a different profile of port staff – those with the credentials needed to access and manipulate automated systems.

To account for such changes, the authors propose using strategic foresight – the systematic exploration of multiple possible futures – and anticipatory governance, which involves the institutional embedding of foresight in policy cycles.

In particular, the report proposes the use of cross-impact analysis, horizon scanning to monitor emerging trends and signals of change, trend analysis and scenario building.

In practice, the authors propose applying this approach to anti-corruption work in three steps:

1. Mapping the system – identifying the actors involved, their interests, incentives and relationships, as well as the political, economic, legal, technological and societal factors shaping corruption risks.
2. Anticipating actor responses and unintended consequences – assessing how different actors may adapt to new anti-corruption measures. The authors also propose identifying in advance indicators that can reveal both emerging risks (red flags) and signs of positive change (progress markers).
3. Scanning the horizon for trends, new risks and opportunities – tracking broader political, economic, legal, technological and societal shifts that may create new corruption risks or, conversely, open up opportunities for reform.

A sign that corrupt or criminal networks are adapting may be the absence of the expected effect of an intervention. For example, if stronger enforcement against drug trafficking does not lead to higher drug prices, this may indicate that supply remains unaffected because traffickers are using counter-strategies. Possible signs of positive change identified by the authors include increases in whistleblowing, voluntary compliance and public trust.

The report also pays particular attention to the use of artificial intelligence. AI tools can help scan large volumes of information for signals of corruption risk, generate and stress-test scenarios, model how actors might respond to anti-corruption interventions and organise analytical results. At the same time, the authors stress that technological tools cannot replace expert judgement and contextual knowledge.

Based on the study, the Basel Institute recommends shifting from linear to adaptive approaches to the design and implementation of anti-corruption interventions, adopting a complexity lens from the outset, incorporating foresight and anticipatory thinking into the design and monitoring of anti-corruption measures, and tailoring the scale of such work to the context and the resources available.

At the same time, the report emphasises that these methods are not intended to predict future corruption with certainty: an element of unpredictability will always remain. Rather, they provide a more systematic way of thinking about uncertainty, adaptation and change, allowing policymakers and anti-corruption practitioners to consider in advance how corrupt practices may evolve and to adjust interventions accordingly.