Taylor Swift Deepfake Issue: Microsoft brings new security measures to its AI tools

Microsoft has introduced more protections for Designer, its artificial intelligence text-to-image generation tool that allows users to create non-consensual sexual images of celebrities.

According to 404 Media, AI-generated nude photos of American singer and songwriter Taylor Swift went viral last week on 4chan and Telegram channels, where people use Designer to create AI-generated images of celebrities. .

“We are investigating these reports and taking appropriate action to resolve these issues,” a Microsoft spokesperson said.

“Our Code of Conduct prohibits the use of our tools to create adult or non-consensual intimate content, and any repeated attempts to create content that violates our policies may result in a loss of access to the service. We have a large team dedicated to operating in accordance with our responsible human Smart principles develop guardrails and other safety systems.”

Microsoft said that its ongoing investigation cannot confirm whether the Swift image on X was created using Designer. However, the report states that the company is continuing to strengthen its text filtering tips and address abuse of its service.

Meanwhile, Microsoft Chairman and CEO Satya Nadella said the apparent Swift AI fraud was “shocking and horrific.”

“I think we need to move quickly on this,” Nadella said in an interview with NBC Nightly News. Swift is reportedly considering taking legal action against the website responsible for generating the deepfakes.

(This article has not been edited by News18 staff and is published by IANS, the associated news agency)