While expense fraud itself is nothing new, the scale, sophistication and accessibility of AI tools have fundamentally changed the landscape. For UK businesses, this represents a growing compliance risk that can no longer be ignored.
Why AI Has Changed the Game
Historically, falsifying a receipt required some level of effort; editing software, manual manipulation, or purchasing templates online. Today, generative AI can produce highly realistic receipts in seconds using nothing more than a short prompt.
These AI-generated documents often include convincing details such as itemised lists, VAT calculations and realistic formatting, making them difficult to spot even for experienced reviewers.
More concerning is how quickly this trend has grown. Data from expense auditing platforms shows AI-generated receipts went from virtually non-existent in 2025 to over 70% of flagged fraudulent documents by mid-2026.
This isn't just a technological shift, it's a behavioural one too. Surveys suggest a significant proportion of employees have experimented with AI tools for expense claims, whether to recreate lost receipts, inflate legitimate costs, or fabricate entirely new purchases.
Why It's Harder to Detect
One of the biggest challenges for businesses is that traditional controls are no longer sufficient.
Manual checks that rely on visual inspection simply aren't robust enough anymore. Modern AI tools can replicate paper texture, realistic imperfections and consistent formatting - all the things finance teams have historically relied on to identify fakes.
There's also a shift in how fraud is being carried out. Rather than submitting one large false claim, employees are increasingly submitting multiple low-value claims designed to stay under approval thresholds.
Individually, these claims may seem insignificant, but collectively they can create a meaningful financial leak.
The Risk to UK Businesses
For SMEs and owner-managed businesses in particular, this presents several risks:
- Financial loss through undetected fraudulent claims
- Compliance issues, especially where VAT is incorrectly claimed on fictitious expenses
- Reputational damage if controls are found to be weak
- Internal trust erosion, as fraud often originates from within the organisation
As AI tools become more widely available, the barrier to committing this type of fraud has effectively dropped to zero, meaning the issue is only likely to grow.
How Businesses Can Respond
The good news is that there are practical steps businesses can take to reduce their exposure:
- Strengthen expense policies
Clear guidance on what can and cannot be claimed is essential. Policies should be regularly reviewed and communicated to staff to remove ambiguity.
- Reduce auto-approval thresholds
Lower-value claims are increasingly being used to bypass controls. Tightening approval limits can help prevent this.
- Link expenses to evidence beyond receipts
Where possible, match claims to supporting data such as calendar entries, travel bookings or company card transactions.
- Implement technology-driven checks
AI is being used to commit fraud, but it can also be used to detect it. Modern expense systems can identify anomalies, inconsistencies and patterns that humans may miss.
- Encourage a culture of accountability
Fraud often thrives in environments where processes are seen as informal or poorly monitored. Reinforcing expectations and accountability is key.
Final Thoughts
AI is transforming business operations in many positive ways, but it is also creating new risks. Expense fraud is no longer just about spotting a poorly edited receipt; it's about recognising that the entire system of verification has changed.
For UK businesses, the message is clear: relying on “what looks right” is no longer enough. Stronger controls, better systems and a proactive approach are essential to staying ahead.