Bitdefender, a cyber security company, shares a report that shows that there are many advertisements on Facebook and Instagram that use celebrities' deepfakes to sell different products on these sites. The voices of many celebrities are played over their clips to offer many product giveaways and convince them for investments. Bitdefender analyzed many videos and found that they targeted more than one million users in the US and Europe and these posts reached 100,000 people on average on different social media sites. Another thing that these scammers did was create fake news websites of popular news sites like the New York Times and link the deep fake posts to those news websites.
Bitdefender’s Security Analyst, Slime Bizga, said that the company hasn’t been able to identify what are the number of scams this has caused but there is a rise seen in them from the past few months. This report was published after Meta has announced that it is going to label AI generated content on its social media sites while working with technical experts to identify AI content. The FCC has also said that it is thinking about banning AI generated voice overs on robocalls after an incident where the AI generated voice of Joe Biden could be heard discouraging New Hampshire voters from going to polling stations.
Bizga said that as elections are near, these kinds of deep fakes of politicians can be seen but her team hasn’t looked into them but it is highly likely that the deepfakes of politicians are as widely being used as celebrities’. Bizga also added that deep fakes of celebrities on social media sites can easily be identified because the voices aren’t edited perfectly. They are not synced, there is something wrong with movements and their quality of voices isn’t great. The deepfake of a youtuber, Mr. Beast, the only one that users weren’t able to identify easily. In future, the deep fakes are going to get more advanced. So, users need to be alert and keep an eye on every content to prevent themselves from being scammed.
Photo: DIW-AIgen
Read next: Research Finds that Many of the Adults in the UK Are Not Aware that AI Generated CSAM is Illegal
Bitdefender’s Security Analyst, Slime Bizga, said that the company hasn’t been able to identify what are the number of scams this has caused but there is a rise seen in them from the past few months. This report was published after Meta has announced that it is going to label AI generated content on its social media sites while working with technical experts to identify AI content. The FCC has also said that it is thinking about banning AI generated voice overs on robocalls after an incident where the AI generated voice of Joe Biden could be heard discouraging New Hampshire voters from going to polling stations.
Bizga said that as elections are near, these kinds of deep fakes of politicians can be seen but her team hasn’t looked into them but it is highly likely that the deepfakes of politicians are as widely being used as celebrities’. Bizga also added that deep fakes of celebrities on social media sites can easily be identified because the voices aren’t edited perfectly. They are not synced, there is something wrong with movements and their quality of voices isn’t great. The deepfake of a youtuber, Mr. Beast, the only one that users weren’t able to identify easily. In future, the deep fakes are going to get more advanced. So, users need to be alert and keep an eye on every content to prevent themselves from being scammed.
Photo: DIW-AIgen
Read next: Research Finds that Many of the Adults in the UK Are Not Aware that AI Generated CSAM is Illegal
