The pellets are being used to care for the school’s White Pekin ducks.

For several months, the ducks have been benefiting from the pellets thanks to a donation from Land Energy, a nationwide biomass manufacturer and distributor.

St Peters school pupils feeding the ducks (Image: St Peter’s CE Primary School)

Emily Lemaire, school administrator, said: “We rallied the community and quickly raised more than £1,000 to buy and build a new duck run for our Forest School area.

“Nearly half of the school is now involved in helping to care for our chickens and ducks, with volunteer ‘duck leaders’ taking part in a day-to-day care rota.

“It can be quite costly to look after the ducks, which is why we were so grateful to Land Energy for donating some of their Sorbeo horse bedding pellets for us to try. The pellets have been fantastic, helping to keep the ducks’ sleeping areas clean, dry and comfortable.”

The ducks in the forest school area (Image: St Peter’s CE Primary School)

The project has engaged a large part of the school community, with many pupils helping care for the animals.

David Bone, communications officer at Land Energy, said: “We are always keen to support schools and community groups that help strengthen our relationship with the natural world, while contributing to our goal of creating real environmental gain and encouraging wider community engagement.

“We are very proud to support St Peter’s CE Primary School and its now three ducks with our Sorbeo horse bedding pellets in what we believe is an unconventional first.”

The Digital Regulation Cooperation Forum has published findings on how four UK regulators are using generative AI in their work. The review covers the Competition and Markets Authority, the Financial Conduct Authority, the Information Commissioner’s Office and Ofcom.

Generative AI is moving beyond pilot projects into regular use across regulatory analysis, supervision and enforcement, the forum found. Joint work over the past year has focused on governance, prompt design, detecting consumer harm and testing AI tools before wider deployment.

The report offers one of the clearest accounts yet of how UK public authorities are trying to use large language models and related systems in day-to-day oversight of digital markets and services. It also shows the extent to which regulators are building internal tools rather than relying solely on off-the-shelf products.

From pilots

According to the findings, each member regulator has been testing, developing or deploying generative AI tools. Staff have also been given access to AI productivity software as part of a broader effort to build familiarity with the technology and set rules for its use.

That work has been supported by six cross-regulator deep-dive sessions involving leaders working on advanced regulatory technology. The sessions examined how to manage risks such as hallucinations and bias, improve outputs through prompt design, identify harmful online design practices and assess whether AI systems are reliable enough for specific regulatory tasks.

A central conclusion is that governance remains critical. AI tools may offer clear benefits, but in a regulatory setting where errors can affect enforcement decisions and consumer protection, they require proportionate oversight, clear accountability and human review.

The Financial Conduct Authority was cited as one example of a regulator using a structured internal framework, including policies on data management, frontier AI and privacy, alongside staff training on risk management systems.

Consumer harm

Some of the clearest examples relate to monitoring websites and apps for harmful design patterns. These include drip pricing, misleading scarcity claims, reference pricing and so-called sludge practices that make it harder for consumers to cancel subscriptions or navigate terms and conditions.

The Competition and Markets Authority has developed what the forum described as agentic AI that can experience and record the consumer journey at scale. The aim is to detect possible breaches of consumer law by navigating online services in a way that mirrors how users encounter prices, prompts and design choices.

This approach has already fed into enforcement activity. The authority has opened investigations into eight businesses and sent advisory letters to 100 others following the investment, according to the forum.

The Financial Conduct Authority has also tested whether large language models can be used for sludge audits. Instead of relying on staff to click manually through websites and recreate customer journeys, the regulator ran a pilot to see whether models could simulate consumer personas and carry out many of those checks more quickly.

The pilot found that the models could perform a large share of the audit work, but with important caveats. Prompt design was needed to improve consistency, the systems did not always interpret webpages correctly and human review remained necessary to check accuracy.

Ofcom has been using behavioural audits in its own work to examine whether online services meet obligations under the Online Safety Act. These audits look at areas such as sign-up processes, features that influence time spent on a service, negative sentiment tools and reporting mechanisms.

The Information Commissioner’s Office and the Competition and Markets Authority have also worked jointly on harmful design in digital markets, while the Information Commissioner’s Office has monitored compliance on the use of non-essential cookies at scale.

Testing tools

Another strand of the programme focused on how regulators assess the quality of AI systems before wider deployment. Several have created minimum viable internal evaluation frameworks that define the task an AI tool is meant to support, set guardrails for users and test for accuracy, usefulness and obvious failure modes.

These frameworks compare model outputs against reference answers using test questions, source documents and preset criteria such as factual accuracy, substance and citation style. Models are then scored against pass-fail thresholds.

This approach helps technical teams decide whether a tool is suitable for sensitive work and supports a more controlled move from trial use to broader deployment. But the process can be time-consuming because reference materials and benchmark answers must be created in advance.

Prompt engineering emerged as another practical area of work. Regulators reported that better outputs often depend on supplying context, defining a role for the model, giving examples and setting clear constraints. More advanced methods included breaking down complex questions into smaller parts and linking prompts in sequence.

The findings also refer to the use of retrieval-augmented generation, in which models draw on approved internal documents to ground responses in verified material. The forum said this can reduce the risk of hallucinations, but not remove it, so it is being used alongside other checks.

Overall, the picture is of regulators using AI to cut the time and cost of investigations without removing human judgment from the process. In areas such as consumer protection and online safety, the technology is being positioned as a way to monitor large numbers of services and spot patterns that would be difficult to identify manually.

Cross-regulator cooperation has helped speed adoption, reduce duplication and support more consistent approaches across the four bodies, the forum found. Strong governance, evaluation frameworks and human oversight remain essential to the safe and effective use of generative AI in regulation.

Fraud losses in the UK rose to £1.28 billion in 2025, according to FICO, which linked the increase to a shift towards scams that manipulate victims rather than break into accounts.

Figures cited by the analytics group showed that more than four million fraud cases were reported across the UK during the year. The pattern of losses suggests criminals are moving away from attacks that rely on technical access and towards fraud that persuades people to make payments themselves.

James Roche, Fraud Consultant at FICO, said the strongest growth came in authorised payment fraud, where victims approve transactions after being deceived. Losses in that category rose 19% to £576.4 million.

Within that total, so-called malicious payee fraud was a main driver. In these cases, victims believe they are paying for a legitimate product, service or investment.

Investment scams rose 40% to £221.5 million. Purchase scams increased 20% to £118.1 million, while romance fraud climbed 23% to £39.2 million and advance fee scams jumped 65% to £58.4 million.

By contrast, losses from malicious redirection scams fell to their lowest level since 2020. In these cases, criminals impersonate police officers, bank staff, chief executives or procurement staff to divert payments.

“Every scam that requires technical access to an account is declining,” Roche said.

“Every scam that requires psychological access to a person is accelerating. This is because fraud prevention systems were designed to detect unauthorised access. Now they need to identify people who are being tricked in real time.”

Card fraud

Unauthorised card fraud followed a different pattern. The number of card fraud cases rose 11%, but losses were broadly flat, suggesting criminals are targeting larger numbers of victims for smaller amounts.

“Criminals are targeting far more people for smaller sums in this space, perhaps because the perpetrators are less sophisticated fraudsters using compromised cards or stolen cards,” Roche said.

Remote purchase fraud, also known as card-not-present fraud, now accounts for more cases than all other unauthorised fraud categories combined, FICO said. It pointed to social engineering tactics used to obtain one-time passcodes for digital wallet registration as a key factor.

The figures reflect a broader change in how fraud is organised, according to Roche. He said criminals are moving away from one-off or physical attacks and towards broader networks built around exploiting victims directly.

“Criminals are shifting from system compromises to human exploitation and from opportunistic or physical fraud to organised fraud ecosystems,” Roche said.

That shift is also changing what banks and payment providers need from fraud controls. FICO said firms must assess customer intent before a payment is authorised, monitor behaviour in real time and share information more effectively across teams and systems.

Roche said the industry was moving away from isolated tools aimed at single risks and towards shared platforms that combine signals across different products and channels. He added that banks needed to use artificial intelligence more effectively than fraudsters while keeping decisions explainable.

FICO also referred to work involving its Scam Signal service and Jersey Telecom, which uses telecommunications data to flag when a consumer may be under pressure during a live conversation, such as an impersonation scam. Roche said that can give a bank a chance to contact a customer before money is sent.

“We are seeing a real impact from the use of our innovative Scam Signal solution, in collaboration with Jersey Telecom,” Roche said.

“This solution uses telecommunications data to identify when a consumer may be being tricked in a conversation such as an impersonation scam, enabling the bank to reach out to the consumer and thwart the attempt.”

Changing tactics

Roche said the next stage of the response would depend on a fuller view of customer behaviour across channels and products, as well as preparation for fraud threats linked to agent-led interactions and newer prevention technology.

“Fraud teams need a deep cross-channel, cross-product, customer-centric view of everything their customers do, and who they are,” Roche said.

“This is how banks win against the fraudsters. The banks’ advantage is that they can create context and deep-level relationships to protect customers when they need them the most. Fraudsters can’t achieve that; they deploy the same strategies to many, hoping to find the most susceptible people.”