Whilst artificial intelligence (AI) can book us dinner reservations and edit photos with a single command, it can also create eerily realistic deepfakes that blur the line between real and fake. And that’s a big problem. If people can’t trust what they see online, how are they supposed to trust the brands and services they rely on?
Enter Honor, the Chinese tech company that’s taken on the deepfake problem headfirst. Their latest Magic 7 Pro smartphone comes equipped with an AI-driven deepfake detector that claims to verify content authenticity in just six seconds. That’s faster than most of us can decide what to watch on Netflix.
“The whole process is completely running on-device, so you don’t need to worry about privacy - your safety is totally taken care of,” says Honor AI product expert Thomas Bai. In other words, you don’t have to send your data off to some mysterious cloud server to know whether that viral video is real or AI trickery.
Smarter AI that doesn’t cost you your privacy
People want AI to be more intelligent, but they also want their personal information kept private. It’s a tricky balance - one that has left many consumers skeptical about AI’s intentions (and occasionally reaching for their tinfoil hats).
Honor’s solution? Keep it all on-device. By processing deepfake detection locally, their AI tech eliminates the need to send data to external servers, keeping user information secure while still offering powerful AI capabilities.
“The AI agent uses a technology called GUI, which stands for Graphic User Interface,” Bai explains. “It means our AI agent can understand the screen, read it like a human and act on behalf of the user’s permission.”
Essentially, it’s like having a digital assistant with eyes, but one that actually respects your privacy. Now that’s refreshing.
Why customer trust matters more than ever
Deepfake detection isn’t just about preventing people from falling for a fake celebrity endorsement of a dodgy skincare product. It’s a much bigger deal. Consumers don’t just shop online - they bank, invest and seek legal advice digitally. And cybercriminals know this all too well.
Take the case of Martin Lewis, the UK’s beloved money-saving expert. His deepfaked face and voice were used in a scam that promised lucrative Bitcoin investments. One victim, Des Healey, was convinced by a hyper-realistic video of Lewis and Elon Musk promoting the scheme. The result? Healey lost £76,000.
Platforms like Facebook have since pledged to use facial recognition to curb such scams, but the reality is that fraudsters are getting more sophisticated by the day. If even a seasoned financial guru like Martin Lewis can have his identity hijacked by AI, everyday consumers stand little chance without better defenses in place.
Social media and the erosion of trust
Speaking of Facebook, did you know it just turned 20? You’d think after two decades, it would have built up a decent level of trust, but according to the Thales 2024 Digital Trust Index, only six percent of people actually trust social media companies with their data. That number drops to a measly three percent in the UK.
With countless data breaches, privacy scandals and, yes, deepfake scams happening across these platforms, it’s no surprise people are wary. Security experts argue that social media companies need to overhaul their approach to offer more transparency, accountability and user empowerment if they want to regain public confidence.
As Javvad Malik, a security awareness advocate at KnowBe4, put it: “With the continuous number of breaches and misuse of data at social media providers, it is no surprise that people don’t trust these organizations to safeguard their data.”
But what if companies took a different approach? What if instead of mining every detail of your life, they offered tools - like on-device deepfake detection - that actually protected users from scams and fraud? Now that would be a social media revolution worth celebrating.
Consumers want smarter, more intuitive technology, but they also want to feel safe using it. If on-device deepfake detection proves effective, it could set a new standard where AI helps protect people rather than deceive them. So, next time you see a celebrity promoting an investment that seems too good to be true- maybe take six seconds to double-check. Your bank account will thank you.