Article Source: Quantum Bit

Deepfake's latest victim has appeared - global superstar, Taylor Swift.

Almost overnight, social platforms such as X and Facebook have been flooded with a large number of fake "indecent photos," with views already reaching tens of millions!

△ Image synthesized by Deepfake

Although the account that first spread these photos has been banned, the circulation of the photos has not been completely stopped.

Taylor Swift herself has become aware of the situation and expressed considerable anger, and fans have spontaneously organized a protection campaign.

On X, some netizens directly @ Musk, calling on him to come out and take charge of the situation.

Viral spread of fake photos

The incident began with the romantic relationship between Taylor Swift and American NFL football player Travis Kelce.

Kelce plays for the Kansas City Chiefs, and after their relationship was exposed, Taylor Swift also became a regular at the team's games.

So, some people took the opportunity to use AI to generate explicit photos with the stadium as the background, involving Taylor Swift and several fictional "players."

△ Image synthesized by Deepfake

There are various rumors about the spread of these photos, with The Verge claiming that they first appeared in a certain encrypted chat group.

The UK's Daily Mail claims that these Deepfake photos were initially circulated on an illicit website that is filled with a large number of indecent photos featuring celebrities.

In addition to Taylor Swift, this website also contains hundreds of celebrities, including Margot Robbie, who played Barbie.

Shortly after the photos were posted online, they began to spread like wildfire on X, Facebook, and other major social media platforms, reportedly reaching tens of millions of views.

"Taylor Swift AI" briefly became a hot topic on X, but it is now suspected to have been blocked.

After learning of the situation, Taylor Swift's fans launched a protection campaign on X, and "PROTECT TAYLOR SWIFT" became a new hot search.

However, some people infiltrated the campaign and continued to spread those fake photos under the guise of the "PROTECT" tag…

Angry fans uncovered a suspected instigator of the widespread dissemination of the photos, and claimed that the individual has a history of years of sexual harassment.

The account has currently been set to private, but according to users with access, the user arrogantly taunted, "You can't find me," on their account.

As of now, some accounts involved in spreading the photos have been gradually banned by major platforms, but many are still at large.

Some individuals have even openly sold a collection of "self-made" deepfake photos of Taylor Swift on their self-built website, with a price of £10 for a set.

A source claiming to be close to Taylor Swift told the Daily Mail that she is very angry about the situation, and her team is considering legal action.

It is also understood that this is not the first time Taylor Swift has encountered Deepfake, but the previous incident only involved impersonating her image and voice for product endorsements, without explicit content.

A report by the Associated Press shows that there were as many as 140,000 Deepfake videos on the internet last year, surpassing the total from previous years.

In the face of Deepfake, celebrities like Taylor Swift still have fans and teams to help clarify and speak out, but ordinary people who encounter such situations may find it difficult to defend themselves…

Efforts to defend against Deepfake

In this incident, the timely deletion of posts by major platforms may have been effective. On X, most related content is no longer visible.

Regarding Facebook, a spokesperson for its parent company Meta told the Daily Mail:

This content violates our policies, and we are in the process of removing it from our platform and taking action against the accounts that posted it.

However, some netizens argue that this is not just a matter of deletion; these photos should never have appeared on these platforms in the first place.

Meta has responded to this, stating that they "will continue to monitor and promptly remove any violations."

On the technological front, Nature magazine also mentioned Deepfake detection in its article "Seven Technologies Worth Watching in 2024."

The article in Nature points out that incorporating invisible watermark mechanisms into AI tools may be a solution for developers.

However, this approach may be more for tracking after the fact, while also serving as a deterrent. To truly prevent Deepfake at its root, it may require a collective effort from all of humanity…

Reference links:
[1]https://www.theverge.com/2024/1/25/24050334/x-twitter-taylor-swift-ai-fake-images-trending
[2]https://www.dailymail.co.uk/news/article-13006645/taylor-swift-furious-ai-pictures-porn-legal-action.html
[3]https://www.nature.com/articles/d41586-024-00173-x

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。