Microsoft engineer whistleblower alleges he was ‘silenced’ after alerting about DALL-E issue linked to Travis Kelce’s girlfriend Taylor Swift’s n*des

Recently, morphed images of Taylor Swift created with the help of AI resulted in a massive outrage online.


Microsoft engineer whistleblower alleges he was ‘silenced’ after alerting about DALL-E issue linked to Travis Kelce’s girlfriend Taylor Swift’s n*des

Taylor Swift (via People)

The recent deepfake leak of Taylor Swift‘s n*de images and the subsequent circulation of these images on social media sites such as X (formerly known as Twitter) led to the site blocking down searches for the pop star. After some investigation, it has been found that one of Microsoft’s AI programs was used to create these images. This has invited a lot of criticism from her fans and the White House took it seriously and made a statement on it.

An engineer working with Microsoft revealed that this problem had existed for a while. The company knew that OpenAI’s DALL-E program contained security loopholes that could be misused to create images targeting individuals. However, the company silenced him as per a reputed source.

YouTube video

The engineer revealed he discovered this particular issue early on in December 2023 and reported it to the company. He was instructed to reach out to OpenAI, the company which is now a part of Microsoft.

FS Video

That was the last he ever heard of the issue. Nobody bothered reaching out to him with an appropriate response. Tired of this, Shane Jones decided to go rogue and he detailed his criticisms in a lengthy LinkedIn post that has since been deleted as Microsoft’s legal team ordered him to delete the same.

Taylor Swift’s morphed images reveal the potential for misuse of AI-based programs

Unsatisfied with this, he reached out to Congress and the Attorney General of Washington, Bob Ferguson, and wrote a letter warning them about the issues. Ferguson stated,

The vulnerabilities in DALL.E3, and products like Mircrosoft Designer that use DALL E3 makes it easier for people to abuse AI in generating harmful images. Microsoft was aware of these vulnerabilities and the potential for abuse.
Taylor Swift
Taylor Swift (Via Open Source/X)

With AI tools easily accessible and their use more prevalent than ever, the risks associated with them are slowly coming to the surface. Even though it can do wonders, the technology is still in its relatively early stages and, as a result; it is easy to tamper with it and misuse it.

After this debacle, a bipartisan group of US Senators introduced a bill on Tuesday that demands the criminalization of spreading nonconsensual, s*xualized images generated by AI. This is a welcome step, as it would greatly benefit victims who have been wronged as a result of deepfakes to pursue legal action and receive justice. Taylor Swift nor Travis Kelce have addressed this matter or commented anything regarding this situation.

In case you missed it: