On February 14, 2024, Microsoft announced that it was taking action to stop people from using its artificial intelligence to create inappropriate fake images of celebrities. This decision follows an incident where explicit deepfake images of singer Taylor Swift were made using Microsoft’s Designer tool. These fake images then spread widely online.
The deepfakes of Taylor Swift were made using Microsoft’s artificial intelligence software. After the images leaked and went viral, Microsoft decided it had to do something. The company is now blocking its AI from generating fake celebrity pornographic images.
This move by Microsoft aims to curb abuse of its powerful AI tools. The company wants to prevent them from being used to create and spread unethical deepfakes targeting high-profile individuals without consent.
Severity of the problem
We are experiencing a swift increase in malicious individuals’ misuse of advanced artificial intelligence tools. This misuse includes the creation of deepfakes, which are generated using AI techniques to manipulate videos, audio recordings, and images. \
This disturbing trend presents fresh challenges in various areas, such as elections, financial fraud, harassment via non-consensual pornography, and the rise of cyberbullying in the next generation.
The manipulation of media through deepfakes can undermine the integrity of elections by spreading false information or altering the perceptions of candidates and voters.
Financial fraudsters can use AI-generated content to deceive individuals or organizations, leading to significant monetary losses. Additionally, the creation and dissemination of non-consensual pornography through deepfake technology can cause immense harm and distress to victims.
Microsoft’s response
In response to this incident, Microsoft has taken proactive measures to prevent the misuse of its AI imaging tool, Microsoft Designer. It was discovered that the tool was being exploited due to a flaw, allowing users to create content that violated the service’s usage rules.
To address this issue, Microsoft has implemented safeguards and controls to block the unauthorized creation of content that goes against the AI imaging tool’s intended purpose and usage guidelines. By identifying and rectifying the flaw, Microsoft aims to uphold the integrity of its service and prevent any further misuse or exploitation by individuals seeking to circumvent its rules.
“We are investigating these reports and are taking appropriate steps to address them. Our Code of Conduct prohibits using our tools to create intimate adult or non-consensual content, and any repeated attempts to produce content contrary to our policies could result in loss of access to the service. The company said in a statement.