Starling Bank Is Helping You Spot AI Scams, Here’s How

Starling Bank has launched a new tool called Scam Intelligence, designed to help customers detect fraud before sending money to scammers. It is the first UK bank to place an AI-powered scam detection feature directly into its app. The feature analyses images and text from marketplace listings and gives specified advice to user’s individual needs within seconds.

The tool works through Starling’s mobile banking app and is available for personal, joint and business customers. It uses Google’s Gemini AI model, supported by Google Cloud, to read images, understand the context, and assess risk. The analysis takes place securely, with Starling confirming that no banking data is shared or stored within Gemini. The uploaded images are processed temporarily and only for the purpose of checking scam indicators.

Graham Dury, Director, FSI, Google Cloud in the UK and Ireland said: “Fraud is a huge problem for UK consumers, and we believe AI can be a powerful tool to help tackle it.

“By working together with Starling and using our Gemini models on a secure cloud platform, we’re giving customers a new way to spot the signs of a scam. It’s a brilliant example of using AI to help protect millions of people.”

Starling said Scam Intelligence follows the success of its earlier tool, Spending Intelligence, which launched in June. Both tools aim to help users understand their money habits and protect themselves from fraud through technology. Harriet Rees, Starling’s Chief Information Officer, said that knowledge is power when managing money, and AI is giving customers exactly that.

“We’re on a mission to help our customers be ‘Good with money’, from how they spend it to how they save and protect it. While we work with social media platforms and mobile networks to combat fraud at source, Scam Intelligence is a tool people can use to help protect themselves.” said Catherine Britton, Head of Fraud Risk at Starling Bank.

How Does The Tool Help Customers Spot Scams?

Scam Intelligence scans and analyses uploaded images of items, as well as listings or messages from online sellers. Customers can submit screenshots from platforms such as Facebook Marketplace or eBay, and the system will detect possible scam warning signs.

If someone tries to buy a bicycle from Facebook Marketplace, the tool may flag that the price is suspiciously low or that the image looks fake. In another example, if a person is buying a guitar on eBay, Scam Intelligence might identify that the seller refuses to use eBay’s secure payment features. The tool can also recognise signs of pressure, such as messages pushing a buyer to transfer money quickly before seeing an item.

Once the scan is complete, the customer receives feedback explaining what might be wrong with the listing. This gives them a chance to reconsider before making a payment. Starling hopes that such early detection can prevent authorised push payment (APP) fraud, a type of scam where people unknowingly approve transfers to criminals.

Why Is This A Necessary Launch?

Fraud continues to be a costly problem in the UK. According to Starling, scammers cost the country £1.17 billion in 2024. Of that amount, APP fraud accounted for £450 million. These scams have grown more common as online shopping and marketplace trading have become part of everyday life.

Scam Intelligence arrives at a time when online marketplaces like Facebook Marketplace, Gumtree and Vinted have become hotspots for fraudulent listings. Many people fall victim to fake deals or pressure tactics that lead them to send money before verifying sellers. Starling is trying to reduce the number of APP fraud cases linked to its users using AI to warn customers before they transfer money.

The bank also views Scam Intelligence as an educational tool. Over and above stopping immediate scams, it encourages customers to recognise the tricks that fraudsters use. This could make people more cautious and informed in their future online transactions.

How Does Starling Handle Privacy And Security?

Starling said privacy was built into Scam Intelligence from the start. Customers must opt in before using the tool, and all data processed stays within Starling’s secure Google Cloud environment. The images or screenshots uploaded are temporarily shared with Google’s Gemini AI model for analysis, but Starling does not share any personal or financial information.

Once the scan is complete, the uploaded images are deleted and not stored in the system. Starling emphasised that the tool is optional and that customers can choose whether or not to use it. This approach reflects the bank’s effort to use AI responsibly, without affecting customer trust.

What Does This Say About The Use Of AI In Banking?

Scam Intelligence is another move by Starling to make AI a practical part of everyday banking. The bank’s focus on consumer education and fraud prevention makes it unique compared to traditional banks that often rely on “after the fact” fraud detection.

Minister for Fraud, Lord Hanson said: “Fraud is ruthless, and it’s vital that every organisation supports people to avoid scams. I welcome Starling’s new tool, which is a great example of how AI can be used in the battle against fraud.

“This government is stepping up our response, bringing forward an expanded Fraud Strategy to disrupt criminal networks and protect the public. I continue to urge everyone to follow Stop! Think Fraud advice.”