AI-Amplified Virtual Stripping Techniques and the Struggle against Deepfake Exploitation
In the rapidly evolving world of artificial intelligence (AI), a new ethical dilemma has emerged: the use of AI technology to create a bot that digitally removes clothing from images. This bot, which has been used to target at least 100,000 women as of July 2020, with a significant portion suspected to be underage, exposes a vulnerability that malicious actors are quick to exploit.
The bot, available on the messaging app Telegram, allows users to submit images of clothed individuals and receive back manipulated images depicting them nude. This malicious use of AI adds a new layer of complexity to an already pervasive issue: non-consensual intimate imagery.
Several efforts are being made to combat this malicious use of AI-powered "undressing" bots and deepfake technology. One such initiative involves legal action. Meta, for instance, has taken legal action against companies like Crush AI, which circumvented Meta's ad policies to promote AI-generated "undressing" apps. These apps were used to create non-consensual sexual imagery, contributing to sextortion and abuse.
Other social media platforms are also taking steps to regulate their platforms. Meta and other platforms are working to block and remove ads promoting these apps, though the constant shift in tactics by malicious actors makes this challenging.
Legal developments are also playing a role in combating this issue. Argentina has established a legal precedent by criminalizing the creation of child abuse imagery using AI. This ruling aims to combat online child exploitation and sets a model for other regions to follow similar legal actions. The Take It Down Act, enacted in May, encourages companies to better police their platforms against apps that facilitate image-based abuse, leading to increased efforts to remove such content.
Guidelines for responsible use are also being established. Users are advised to use AI undresser tools only on authorized or consented images. Generating non-consensual deepfakes is illegal in many countries, and using these tools responsibly is crucial. Users should also consider privacy and security measures such as using anonymous logins and checking if the tool complies with privacy laws.
Technological solutions are also being developed to detect and remove AI-generated non-consensual content. Sensity AI, for example, is a cybersecurity company specializing in detecting and mitigating the abuse of synthetic media. Technologies like blockchain can also be used to create tamper-proof records of an image's origin, making it easier to identify manipulated content.
Raising awareness about the existence and potential harms of deepfakes is crucial to empower individuals to identify and report such content. High-profile individuals, including celebrities and journalists, have been targeted with deepfake pornographic content as part of smear campaigns or attempts at silencing dissenting voices.
The Cyber Civil Rights Initiative (CCRI) is an organization dedicated to combating online harassment, including the non-consensual distribution of intimate images. Teaching individuals how to critically assess online content and identify potential deepfakes can help mitigate the spread of misinformation.
The use of the bot creates a perverse gamification of harassment, with its ecosystem incentivizing users to target more individuals and share their creations, contributing to a vicious cycle of abuse. However, with continued efforts from social media platforms, legal bodies, and individuals, the malicious use of AI-powered "undressing" bots and deepfake technology can be effectively combated.
Cybersecurity efforts are being bolstered by the development of technological solutions, such as those by Sensity AI, aimed at detecting and mitigating the abuse of synthetic media like the AI-powered "undressing" bots.Through legal action, companies like Meta are taking steps to combat the creation and promotion of such apps, with Argentina setting a precedent by criminalizing the use of AI for child abuse imagery.From general-news reports, it's clear that the malicious use of AI in creating deepfakes poses a significant threat to crime and justice, including cases of sextortion and abuse.To address this issue, organizations like the Cyber Civil Rights Initiative are educating individuals on how to identify and report deepfakes, empowering them to resist the harmful effects of these malicious technologies.
