Skip to content

UK privacy watchdog examining Microsoft’s AI screenshot tool

Microsoft privacy
Image credit: Tada Images / Shutterstock

The UK’s privacy watchdog is in contact with Microsoft over concerns with the firm’s newly announced feature that constantly takes screenshots on users’ devices.

The Information Commissioner’s Office (ICO) will enquire with Microsoft regarding the Recall feature, part of the tech giant’s new AI product line, first reported by BBC News.

Recall is part of Microsoft’s Copilot+ tool, a generative AI assistant. In a recent announcement, Microsoft explained that with Recall, users can ask Copilot to find content previously viewed on the device.

Recall works by taking images of users’ screens every few seconds, providing a set of data that the AI assistant can refer back to.

The prospect of consumer behaviour being constantly recorded has sparked concerns, with an ICO spokesperson saying the companies must “rigorously assess and mitigate risks to people’s rights and freedoms”.

The spokesperson said: “We are making enquiries with Microsoft to understand the safeguards in place to protect user privacy.”

Microsoft said it has implemented security features to ensure privacy protections. The feature can be customised to only apply to certain apps or websites, it does not take shots during private browsing on Microsoft Edge, it must be opted in to and the company has insisted the images are encrypted and only stored on the device itself.

Jake Moore, a cybersecurity advisor at ESET, told UKTN: “Enabling a feature which has the ability to capture screen data not only offers even more data to the company behind the software but also opens up another avenue for criminals to attack.”

Moore said regardless of the benefits of the feature, a balance “must be kept regarding functionality versus privacy”.

Moore added: “Creating and storing more private data seems unnecessary when cybercriminals continually look for any given vulnerability to exploit.”

This week, Microsoft-backed AI firm OpenAI faced its own backlash after launching a voice for its latest version of ChatGPT that sounded eerily similar to actor Scarlett Johansson.

Johansson said in a statement she had been approached several times to provide her voice for the project, in reference to her role as an AI assistant in the 2013 Spike Jonze film Her.

Read more: Sainsbury’s enters generative AI partnership with Microsoft