By Jonah Ellin
Over the last year, the emergence of deepfake technology has become a significant threat to businesses operating in today’s digital ecosystem. Deepfakes are AI-generated media that can convincingly manipulate audio, video, or images to present false information or make it appear as though someone said or did something they did not, A recent example of this is on the political stage where UK Home Secretary James Cleverly warned of UK election interference with the threat of rigging by Britain’s enemies using AI deepfakes.
If left undetected, these deceptive creations pose a significant threat to businesses, as they have the potential to erode consumer trust, consequently tarnishing the reputation that businesses have worked hard to establish.
Deepfakes can be used to facilitate various forms of fraud, including identity theft and financial scams. In some cases, fraudsters use deepfakes to impersonate an employee or executive at a company to trick others into divulging sensitive information or transferring personal funds, as we saw earlier this year when a finance worker in the US paid out $25 million after a video call with a deepfaked ‘chief financial officer’. Examples like this show just how sophisticated and dangerous deepfakes have become.
Businesses may inadvertently fall victim to these fraudulent activities, resulting in significant financial losses and legal liabilities. Recent scams such as deepfake impersonations of Taylor Swift inviting fans to enter a fraudulent competition for Le Creuset cookware pose a wider issue, negatively impacting the celebrity and product brands in addition to the consumer harm of the scam and the personal damage of having your image stolen.
To reduce these risks, companies must begin to take proactive steps. Early detection of deepfakes is critical to preventing fraud and minimising the damage to a company’s brand reputation. Businesses can start implementing specialised software that can detect deepfakes by analysing minute details such as facial expressions, speech patterns, and lighting. This technology can help to identify deepfakes before they can spread more widely and damage brands to an irreversible extent.
Ensuring regulatory compliance is also critical. Businesses must ensure that they are following regulations related to data privacy and security, which can differ between countries and certain jurisdictions, alongside this, implementing copyright laws is also important to the handling of unauthorised use of content. Enforcing personality rights laws and strengthening platform responses could offer another defence against the spread of deepfakes.
What happens if you or your business ‘deepf’
The current sophistication of deepfakes and how easily they can be produced is unsettling. Deepfakes have now gotten to a stage where even educated viewers may not immediately realise that the content isn’t authentic. Thanks to social media, the risk has gone up considerably; the quality and reach of a deepfake ‘campaign’ are now at levels where the images and videos can have a long-lasting impact, even when disproved.
As with many illegal scams and counterfeiting efforts, at this time there is currently no way to entirely avoid the problem. Platforms that host content have successfully defended in most cases that they are only platforms and therefore not the only ones responsible for the content they host. Most have taken important steps to identify offending posts and remove them as well as to provide steps for users to request the removal of content.
However, identifying deepfakes can take up a lot of time, as platforms are not equipped to deal with deepfakes. Confirming whether someone would have said or done what’s being shown is too complex for the current world of AI – in fact, human behaviour may still surprise people.
What are the first steps you should take if you fear it might have happened to your business?
If you or your brand falls victim to a deepfake, it’s important to take certain steps to disprove the fake version and prevent its spread. Since the internet has a long memory, it’s crucial to make it difficult for others to access and share the content. One way to do this is to request that search engines remove the content from their index. If the content is hard to find, it’s less likely to spread quickly.
The next step is to cut off access to the content. Major social media platforms also have search functions, which is where many people source their news. It’s important to request the removal of the content and any associated posts from these platforms. However, this can be challenging since users can post, share, and repost content even once the original post has been removed, which can multiply quickly. Therefore, it’s best to start by tackling the original post and then move on to the reposts.
The third step is to remove the sources of the content. Videos and pictures often find their way to websites that host this type of content, and these sites may link to posts on social media platforms. Removing such websites can have a broader impact since it leads users to the content. However, it can be difficult to remove content from such websites, and expert or legal advice should be sought in more complex cases.
Final Thoughts
Knowing how to detect and stop deepfakes is essential for maintaining brand integrity. With better tools and technology being introduced to the market at all times, businesses must continue to prioritise authenticity and transparency in the digital realm while also remaining vigilant and implementing effective detection systems to minimise the risks associated with deepfake manipulation.
By taking proactive measures and utilising specialised technology and expert brand protection professionals, businesses can reduce the likelihood of fraudulent activities and protect themselves against the rapidly mitigating risks, potential financial and legal consequences of deepfakes.
About the Author
Jonah Ellin has nearly three decades of experience in product management and technology, and now holds the position of Senior Vice President of Product Management at Corsearch. Leading a team focused on delivering product vision, R&D, and leveraging AI for brand protection solutions, including the innovative Corsearch Zeal platform to combat online counterfeiting.
