News

An interesting editorial blog from a UK legal firm on whether use of AI in internal company processes can increase risk of prosecution under the Failure to Prevent Fraud Act.

This UK Act enshrines a legal principle that is common to other anti-fraud law around the world.  Unlike traditional corporate criminal attribution, the prosecution does not need to establish that any senior individuals within the company knew about or were party to the fraud. Liability arises solely from the fact that a fraud offence was committed (by any “associated party”) that would benefit the company.

The defence is that a company took reasonable steps to mitigate fraud risks, including by its own employees and suppliers.  This is analogous to the “due diligence” defence in food safety law.

The argument here is that you cannot “outsource” this due diligence to AI.  This is because it is foreseeable that AI results will sometimes be incorrect.  In fact, if you use AI without human verification, you are weakening the due diligence defence and foreseeably increasing the likelihood of giving misleading statements.  This applies to relying on AI for everything from supply-chain risk assessments to drafting of company financial reports.

Photo by Tingey Injury Law Firm on Unsplash

Italy on April 15 gave final approval to a food and agriculture protection bill that creates new crimes, tightens penalties tied to company revenue and expands controls across the supply chain..

The legislation also strengthening administrative sanctions and coordination among inspectors.  It creates a new offense of food fraud.. The law also creates a separate offense for the trade of foods with false signs, a category designed to catch misleading labels.

The bill adds specific aggravating circumstances that warrant increased penalties. Among them is “agropiracy,” a term used for organized and systematic illegal activity in the food sector. Penalties are linked to company turnover.

The law also increases penalties for counterfeiting PGI and PDO designations. It also strengthens traceability requirements, with tighter rules on how products are identified and monitored

To improve enforcement, the law creates a coordination body for inspections.

More details are in this media report.

New front-of-pack labelling requirements are being introduced in the US.  This will introduce a motivation for fraud which already exists in many other countries with similar compulsory traffic light systems: deliberately omitting or under-declaring a “bad” ingredient or additive in order to make the front-of-pack summary look “healthier”.

The US “Transparency, Readability, Understandability, Truth, and Helpfulness” (TRUTH) in Labelling Act was introduced last month and would require FDA’s proposed rule regarding front of package nutrition labelling (90 FR 5426 (Jan. 16, 2025)) to be finalized within 180 days of the bill’s enactment.

A principal display panel must identify foods with high amounts of added sugars, sodium, and saturated fat.  High amounts will be based on Daily Reference Values (DRVs). The phrase “High in” and a conspicuous exclamation point icon would be required.

The front of pack panel must also declare the presence of non-nutritive sweeteners and a “factual” statement that such sweeteners are not recommended for children. The wording of this statement has still to be defined, is contentious, and may be dropped from the final version.

Source: Keller and Heckman blog on the Lexology platform.

Photo by Tsvetoslav Hristov on Unsplash

The UK government has issued guidance on the new Failure to Prevent Fraud corporate offence, which is due to come into force 1 September 2025.  This offence has the same “due diligence” defence principle as UK food safety law: if a company cannot show that they have a reasonable fraud defence/mitigation process in place then they become liable if an associate (which could be an employee or a contractor) commits a fraud offence to the company’s benefit.  Their mitigation procedure must be based on the principles typically espoused in food fraud mitigation best practice: they must show evidence of top level commitment, risk assessment, proportionate preventative measures, communication, training, monitoring and review.

The guidance clarifies a number of points, including the territorial scope (for example, a non-UK company can be prosecuted if a UK employee commits the fraud), and that the company can still be liable even if the associate is not prosecuted for the offence (provided that the offence can be proven to the standard required by a court of law).