Use Of Undressing AI Increases, Invites Threats To Privacy And Security

AI Undressing, a new threat in deepfake style. (photo: dock. pexels)

JAKARTA - Social media analysis company, Graphika, revealed that the use of "AI addressing" or "AI is clothinging" is increasing.

This practice involves the use of a generating artificial intelligence (AI) tool that is tailored with precision to remove clothes from images provided by users.

According to their report, Graphika measures the number of comments and posts on Reddit and X containing a referral link to 34 websites and 52 Telegram channels providing synthetic NCII services, and the total is 1,280 by 2022 compared to more than 32,100 so far this year, representing an increase in volume by 2,408% on an annual basis.

Synthetic NCII services refer to the use of artificial intelligence tools to create Intimate Images Without Approval (NCII), often involving creating explicit content without individual consent.

Graphika stated that this AI tool makes it easier for many providers to produce realistic explicit content on a large scale at an efficient cost.

Without this provider, customers will face the burden of managing their own custom image diffusion model, which is time-consuming and expensive.

Graphika warns that increasing the use of AI addressing tools could lead to the creation of false explicit content and contributing to issues such as targeted harassment, sextortion, and child sexual crime (CSAM) material production.

Although AI addressing usually focuses on images, AI has also been used to make video deepfakes using the resemblance of famous characters, including YouTuber Mr. Beast and Hollywood actor Tom Hanks.

In a separate report in October, the UK-based internet watchdog firm, the Internet Watch Foundation (IWF), noted that they found more than 20,254 images of child abuse on a dark web forum in just one month. The IF warns that child pornography generated by AI could "destroy" the internet.

Due to advances in the generative AI image, IWF warns that distinguishing between deepfake pornography and authentic images has become more difficult.

In a June 12 report, the United Nations said that the medium produced by artificial intelligence is a "serious and urgent" threat to information integrity, particularly on social media. Negotiators of the European Parliament and Council approved rules governing the use of artificial intelligence in the European Union on Friday, December 8.