Taylor Swift is not the first victim of AI: decoding the deepfake dilemma

When sexually explicit deep fakes of Taylor Fast went viral on X (Previously known as Twitter), millions of her Fans came together has bury THE AI pictures with "Protect Taylor Fast" posts. THE move work, but he could not stop THE news Since struck each major exit. In THE subsequent days, A fulfilled conversation about THE night of deep fakes was in progress, with White Home press secretary Karine Jean Pierre call For legislation has protect people Since harmful AI content.

But here is THE agreement: while THE incident involving Fast was Nothing short of alarming, he East not THE First of all case of AI-powered content harm THE reputation of A celebrity. There to have has been several instances of famous celebrities And influencers be target by deep fakes on THE last little years – And It is only while going has get worse with time.

"With A short video of yourself, You can Today create A new video Or THE dialogue East led by A scenario – It is amusing if You to want has clone yourself, but THE inconvenience East that someone other can just as easily create A video of You diffusion disinformation And potentially inflict reputational harm," Nicos Vekiarids, CEO of Attentive, A business building tools For validation of Pictures And videos, said VentureBeat.

As AI tools able of create deep fake content continue has proliferate And become more advance, THE the Internet East while going has be buzzing with misleading pictures And videos. This begs THE question: how can people identify what is this real And what is this no?

V.B. Event

THE AI Impact Tour – New York

We will be In New York on FEBRUARY 29 In Partnership with Microsoft has discuss how has balance risks And rewards of AI applications. Request A invite has THE exclusive event below.

Request A invite Understanding deep fakes And their wide range harm

A deep fake can be describe as THE artificial image/video/audio of any of them individual created with THE help of deep learning technology. Such content has has been around For several years, but he begin manufacturing securities In late 2017 When A Reddit user appointed “deepfake” begin sharing AI-powered pornographic pictures And videos.

Initially, these deep fakes largely turned around confront exchange, Or THE resemblance of A person was superimposed on existing videos And pictures. This took A plot of treatment power And specialized awareness has TO DO. However, on THE pass year Or SO, THE increase And propagated of text based generative AI technology has given each individual THE ability has create almost realistic manipulated content – portrait actors And The politicians In unexpected manners has deceive the Internet users.

"It is on has say that deep fakes are No longer THE kingdom of chart artists Or the Pirates. Creation deep fakes has become incredibly easy with generative AI text in photo executives as SLAB, Half-way, Adobe Firefly And Stable Diffusion, which require little has No artistic Or technical skill. In the same way, deep fake video executives are socket A similar approach with text to video such as Track, Picture, On video, Did you see, etc," Vekiarids explained.

While most of these AI tools to have bodyguard has block potentially dangerous instructions Or those involving famous people, malicious actors often figure out manners Or flaws has bypasses them. When investigate THE Taylor Fast incident, independent technology news exit 404 Media find THE explicit pictures were Go on...