Fraud targeting public sector programs has entered a new phase. Sophisticated fraudsters, armed with AI, are depleting public sector budgets and damaging trust in government – and that’s a reality governments must face.

What was once largely opportunistic and fragmented has become organized, industrialized and increasingly cross-border. Fraudsters no longer exploit a single benefit, tax or procurement system in isolation. Instead, criminal networks coordinate attacks across programs, agencies and jurisdictions, deliberately exploiting data silos, policy gaps and slow-moving controls.

For tax authorities, social benefits agencies and public sector finance teams, the challenge in 2026 is no longer simply detecting fraud. It is recognizing how fraud has fundamentally changed and adapting operating models, tools and collaboration in response.

From opportunistic abuse to organized crime

Traditional public sector fraud controls were designed for a different era. They assumed that bad actors were isolated, relatively unsophisticated individuals. That assumption no longer holds.

Today, fraud is often coordinated, repeatable and scalable. Organized groups plan their efforts, pool their resources and divide the labor, frequently replicating successful tactics across programs and borders. The same individuals may appear in social benefits claims, tax filings, or as service providers, but hidden behind different identities.

Structural opportunity allows this criminal coordination to persist. When fraud spans programs, agencies and jurisdictions, detection depends on whether the government can connect the dots. Too often, the connection is buried within individual records and activities. Fraud thrives because it is a lucrative business designed to be undetected.

This reality sets the stage for the macro trends shaping public sector fraud.

Trend 1: Fraudsters exploit fragmentation inside government

One of the most significant enablers of modern public sector fraud is organizational fragmentation.

Government entities often operate with separate systems, data models, and policy mandates. Tax, social benefits, workforce and procurement teams may all hold valuable intelligence, but rarely see the full picture. Fraudsters understand this and deliberately operate across program boundaries, knowing that no single team has end-to-end visibility.

Investigators often detect fraud only after losses occur, relying on patterns surfaced by whistleblowers, audits or post-payment reviews. By then, funds may already be unrecoverable.

Trend 2: Volume and speed overwhelm manual controls

The public sector processes record volumes of transactions, from tax returns and benefits payments to grants, subsidies and relief funds. Manual reviews and rules-based controls cannot keep pace with this pace.

As a result, agencies that rely on manual controls face a growing trade-off: Tighten controls and accept slower service delivery and investigation backlogs, or maintain the same pace and accept the risk of investigator burnout and greater exposure to fraud.

Fraud networks exploit this tension. They move quickly and adapt faster than static controls can respond. By the time patterns are identified, the same tactics may already be active in another program or jurisdiction.

Trend 3: Fraud blends with waste, abuse and non‑compliance

Another defining trend is the blurring of the lines between fraud, waste and abuse.

Many fraudulent activities now sit in grey areas, exploiting policy loopholes, misrepresenting eligibility or manipulating documentation rather than blatant theft. This makes detection harder and increases pressure on frontline staff, investigators and auditors to distinguish intent from mistake.

Important operational questions confront the public sector:

• Which cases require investigation, referrals, prosecution or other actions?
• How do we prioritize limited investigative resources?
• How do we maintain fairness in our decisions while protecting public funds?

If an agency relies on more controls instead of improving risk segmentation and gaining earlier insight, it will fall short.

Trend 4: Technology changes both sides of the fight

Technology cuts both ways. Fraudsters are exploiting data and AI tools to benefit their operations. They harvest data breaches to steal identities and claim benefits for which they would otherwise be ineligible. They create fake documents and photos using generative AI to support their submissions. At the same time, public sector agencies have access to far more powerful data and AI tools than ever before, from advanced analytics and machine learning to network analysis and continuous monitoring.

For agencies considering a shift from isolated fraud projects to ‘enterprise ‑wide’ fraud risk management. This means:

• Connecting data across programs and agencies.
• Shifting from reactive investigations to proactive risk detection.
• Embedding fraud controls into operational processes, not just audits.

This approach can provide agencies with a shared understanding of risk across the organization and enable early action, when intervention is most effective.

What this means for public sector leaders

Fraud can no longer be treated as a back-office issue or the responsibility of a single team.

For any public organization that has financial transactions, the priorities are clear:

• Recognize that fraud is now organized and crosses programs and borders.
• Break down silos that hide patterns and repeat offenders, preventing them from evading detection.
• Equip operational and investigative teams with powerful tools that enable timely, usable insight.

The agencies that succeed in balancing speed, access and integrity without compromising public trust will be those that evolve their operating models as fast as the threat landscape itself.

Read the e-book Gone missing: Can AI save public funds from fraud, waste and abuse?

This blog was originally published on Global Government Forum