Dialling into a video call and wondering if the person staring back at you is actually real sounds like pure paranoia, but realistic deepfakes have moved fast enough to make it a headache.
When a quick digital disguise can mimic your boss’s voice or a client’s mannerisms well enough to authorise a dodgy bank transfer, relying on your eyes alone stops working. To push back against the trickery, a new feature for Zoom aims to stick a verified human badge right beside a caller’s name.
The idea is to prove there’s flesh and blood in front of the lens without forcing software to guess whether individual pixels look suspicious. Getting that tick of approval isn’t just a simple toggle in your settings, though, as it relies on identity checks that ask a fair bit from the person logging in. Knowing what the system actually demands from you, alongside keeping a couple of practical habits up your sleeve, makes it a lot easier to spot a fake before you hand over any sensitive information.
Deepfakes on video calls are a worry these days.
A deepfake is a fake video or voice made with artificial intelligence that can look and sound like a real person. As the tools get better and cheaper, criminals can use them to pretend to be a colleague, a boss, or a supplier on a live call. The aim is usually to trick someone into sending money, sharing private information or approving something they should not.
The problem is bigger than it may seem. A recent study of senior cybersecurity leaders found that 36% had faced at least one deepfake incident during a video call in the past year. That’s more than one in three, which shows these scams are no longer rare.
What the new ‘Human Badge’ does
A company called World has launched a set of tools to help businesses tell real people from AI fakes. The best known is World ID for Zoom, which is now open to business customers. When someone has been verified, a Human Badge appears next to their video feed. The badge is meant to give everyone else on the call more confidence that they are talking to a human. For a manager about to approve a large payment or share sensitive plans, that could be a useful signal.
How the verification works
Getting a badge takes more effort than ticking a box. Workers must first visit one of the company’s physical Orb devices, which scans their iris, the coloured part of the eye, to create a secure reference record. The system uses strong digital protection, known as cryptography, to keep that record safe.
Later, when a person joins a Zoom call, a selfie taken through a mobile app is checked against that reference. If it matches, the Human Badge appears. In short, the system checks that the live person is the same human who was scanned earlier.
What the iris scan means for privacy
Scanning part of your body is a big step, and many people will want to know more before agreeing. Questions to ask include where the data is stored, who can see it, how long it’s kept and what happens if you leave your job. These are fair things to ask any company that handles personal details.
For businesses, the sensible approach is to explain the process clearly to staff and let them ask questions. Taking part may need to be a choice rather than a demand, depending on local rules. People are more likely to accept a new security step if they understand why it’s needed.
The cost and effort for companies
Large firms that want to use the system can choose to have an Orb installed at their own offices, which makes it easier for staff to be scanned. This option is likely to be costly, though, and may suit only the biggest organisations. Smaller teams may find it hard to justify the expense.
There’s also the time it takes. Every person who needs a badge must be scanned and set up, which could be difficult for a large or spread-out workforce. Companies will need to decide which roles and meetings need this level of checking, such as those involving money or sensitive information.
Why proving someone is human may beat spotting fakes
Spotting deepfakes is a never-ending race. As soon as detection tools improve, the people making the fakes find ways around them, and the cycle starts again. The company behind the new tool says this arms race is the reason it chose to prove who is human instead.
The idea is simple. Instead of trying to catch every fake, check that a living person is actually there. It’s a different way of looking at the problem, though it still depends on people being willing to sign up and on the system being hard to trick.
What other tech firms are doing
World is not alone in trying to tackle the problem. Microsoft has been rolling out a series of updates for its Teams meeting app, including opening the platform to outside deepfake-detection tools. The aim is to help stamp out fake video and audio in meetings.
That suggests the big names in video calling all see the same threat. We can expect more badges, warnings, and checks to appear over the coming months. Some will rely on spotting fakes, while others, like this one, will rely on proving who is real.
How to protect yourself on video calls today
You do not need an iris scan to take sensible precautions. If a call includes an unusual request, especially for money, passwords or private files, stop and check through a different route, such as ringing the person on a number you already know. Be wary of any pressure to act fast, since scammers rely on urgency.
Look out for small oddities, such as strange lighting, blurry edges around the face, mouth movements that do not match the words or a video that freezes at awkward moments. Agree a code word with close family or team members for emergencies, and report anything suspicious to your IT team. A short pause to double-check can save a great deal of trouble.
