AI And The New Age Of Exploitation
Author: LawSphereX
Category: Technology, Cyber & Privacy
Published:
Updated:
AI AND THE NEW AGE OF EXPLOITATION
Aasma Baraiya, Institute of Law, Nirma University
ABSTRACT
The emergence of AI and technological advancements has broadened the horizon and revolutionized many industries. However, these advancements are a double-edged sword. Their onset has brought about a new set of challenges with legal and ethical implications. New frontiers of exploitation have emerged, those which are shaped by the power of artificial intelligence. From hyper-realistic deepfakes to AI-generated child exploitation material, sextortion, and human trafficking schemes, the fusion of AI, technology and digital media has unleashed a wave of a unique set of crimes that blur the lines between reality and fabrication. These crimes, often behind the veil of anonymity, challenge traditional legal frameworks, leaving victims vulnerable and justice elusive.
This paper delves into the dark underbelly of technology and AI-driven exploitation, exploring the sinister potential of tools once created for unlocking unprecedented technological possibilities. It examines how deepfake pornography, technology-powered sexual exploitation networks and hidden identities are redefining the landscape of cybercrime. Through a global lens, the research examines real-world instances and implications of misuse of AI and technology, presenting a case study on the infamous New "Nth Room" scandal, which caused a major outcry not only in South Korea, but also garnered international attention due to the magnitude of the crimes committed. Amidst the chaos, hope lies in harmonizing law, technology and social action.
This paper also considers the ways through which society can reclaim control and build a safer digital future—one where technology empowers rather than exploits, harnessing AI’s power to combat exploitative cybercrimes.
1. INTRODUCTION
Artificial Intelligence (AI) has brought transformative changes to industries, unlocking new levels of efficiency, innovation and productivity. However, these advancements have also introduced complex challenges, particularly in the realm of cybercrime. The rise of AI-enabled crimes—from deepfake pornography to AI-facilitated human trafficking—underscores the dual nature of technological progress. While offering unparalleled opportunities for good, AI also provides powerful tools for exploitation. This paper investigates the misuse of AI in cybercrimes, focusing on the gaps in existing legal frameworks and ethical norms. By analyzing real-world examples and exploring the societal impact, this research aims to identify strategies for mitigating the harmful effects of AI misuse and ensuring that technology empowers rather than exploits.
2. AI AND CYBERCRIME: THE DARK SIDE OF TECHNOLOGY
a. Deepfakes and Hyper-Realistic Content The rise of deepfake pornography is among the most concerning examples of ai-driven exploitation. Deepfake technology has been used as a weapon to make non-consensual pornography through the use of AI algorithms to create hyper-realistic but entirely fabricated pictures and videos. By simply layering the faces of innocent individuals onto pornographic material, perpetrators can produce convincing videos and pictures that seem to show real people engaging in sexual activities. 1 This form of exploitation has serious psychological and social costs in addition to violating the victim's dignity and privacy. 2 The easy access of AI tools and the shield of anonymity offered by the internet has enabled the increasing popularity of deepfake pornography.
Anyone with basic technical skills can create a deepfake video with only a few pictures of a person's face. Due to the ease of production, nonconsensual deepfake content is being commonly created and circulated, particularly targeting women and public figures3. In addition to serious emotional distress and reputational damage, victims of deepfake pornography often face threats to their physical safety. 1 https://www. dazeddigital. com/science-tech/article/55926/1/inside-the-disturbing-rise-of-deepfake-porn 2 https://texta. ai/blog/ai-technology/the-rise-of-deepfake-porn-exploring-the-controversies-and-consequences 3 https://www. dazeddigital. com/science-tech/article/55926/1/inside-the-disturbing-rise-of-deepfake-porn
Legal action against deepfake pornography has been weak and inconsistent. Many nations lack specific legal provisions for addressing the issue of deepfake content, however some nations are considering legislation to punish its creation and distribution. Even in nations with such laws, enforcement is still difficult since it is hard to identify and prosecute those who operate anonymously online. Consequently, victims of deepfake pornography often have only limited legal remedies. b. AI-Generated Child Sexual Abuse Material (CSAM) AI-generated CSAM refers to child exploitation images or videos created using AI tools. There are three primary ways AI is used to generate CSAM:
1. Deepfake Manipulation: Existing images or videos of children are altered using AI to
depict them in sexual situations.
2. Text-to-Image Generation: AI models create entirely new CSAM based on textual
descriptions.
3. Nudification Tools: Some AI programs allow users to "remove" clothing from images
of children, creating false but explicit imagery. Unlike traditional CSAM, these images can either be entirely fictional and depict non-existent children or they can be altered versions of real images. 4 This makes it more difficult to differentiate between genuine child abuse material and AI-generated material, thus making it harder for law enforcement to track offenders and victims. The prevalence of AI-generated CSAM has increased rapidly due to easy access to AI tools, through which perpetrators are producing, distributing and even selling CSAM with minimal effort. Other than that, dark web forums and encrypted messaging platforms also enable the anonymous sharing of AI-generated CSAM, making enforcement difficult.
Furthermore, exposure to AI-generated CSAM can desensitise individuals to the horrors of real child sexual abuse, potentially leading to increased demand for such content. 5 4 https://www. cameraforensics. com/blog/2024/11/18/a-guide-to-ai-generated-csam-for-investigators-of-onlineexploitation/
5 https://ssslearning.co.uk/safeguarding-articles/AI-generated-child-abuse
A recent report from the National Center for Missing and Exploited Children (NCMEC) found
that AI-generated CSAM cases rose sharply in 2023, with nearly 5,000 reported instances.6 c. Sextortion and AI-Enhanced Human Trafficking
1) Sextortion
It is a form of blackmail in which perpetrators use explicit images or videos to coerce or threaten victims. 7 Through this, they usually demand money, further explicit content or exploit victims for various illegal activities. Sextortion primarily affects vulnerable groups, including minors, young adults, and women, although anyone can be a target. The rise of deepfake technology exacerbates this issue. Perpetrators can create realistic but entirely fabricated explicit images or videos using AI. Victims may not have engaged in any compromising activities, yet they find themselves blackmailed with seemingly genuine material. This new dimension of sextortion makes it even more challenging for victims to defend themselves and for law enforcement to track down offenders.
According to the National Centre for Missing & Exploited Children (NCMEC), the number of global sextortion cases reported in 2023 were more than double when compared to the previous year, rising to 26,718 compared to 10,731 the year before. 8
2) AI-Enhanced Human Trafficking
Human trafficking is a longstanding global issue, but AI has significantly enhanced traffickers’ capabilities. AI-driven tools are being exploited in multiple ways. AI-powered chatbots and social media algorithms help traffickers identify and groom potential victims. By analyzing online behaviour, traffickers can tailor their approach to manipulate individuals into compliance. Also, advanced facial recognition allows traffickers to track victims, even if they attempt to escape. Furthermore, AI-generated deepfake content can be used to create fake identities, forge legal
6 https://www.missingkids.org/blog/2024/generative-ai-csam-is-csam
7 https://legal-wires.com/lex-o-pedia/understanding-sextortion-forms-elements-and-legal-context/
8 https://www.nationalcrimeagency.gov.uk/news/nca-issues-urgent-warning-about-sextortion\
documents or threaten victims into silence. This makes detection and intervention significantly more complex. AI enables the anonymity of traffickers by encrypting their transactions and communications, making it difficult for law enforcement to track their activities.9
3. EXAMINING THE REAL WORLD IMPLICATIONS THROUGH A CASE
STUDY
The New “Nth Room" Scandal In August 2024, South Korean authorities uncovered over 400 schools, even some of the top Ivy League universities in South Korea involved in running chatrooms where students, primarily middle and high schoolers, participated in the creation and distribution of ‘deepfake pornography’. These so-called "humiliation rooms" targeted their own family, acquaintances and even teachers. Using AI tools, the perpetrators superimposed faces onto sexually explicit content and circulated them across chatrooms. The anonymity provided by online platform further enabled these young individuals to engage in cyber-criminal offences. The way to get into these chatrooms is that a person submits a set of photos of any girl or woman. It has to be someone they know personally.
Then they have to provide personal details of the person in the picture, such as their name, date of birth, occupation, school, grade, age, social media account, phone number and if possible, a home address. They need to provide 10 pictures of them to get access to all explicit deepfake material in the chatroom. Members of the group chat will then make deep fake videos using those pictures and circulate them. 10 There have been many instances of the chatroom members hunting the victims down and showing them their deepfake explicit content. This was done to try to blackmail them into doing explicit activities or to extort money from them. This phenomenon is also known as ‘sextortion’. 11 As news of the scandal broke, panic spread among individuals who feared their images might
9 https://www.hrw.org/news/2024/07/01/using-ai-fight-trafficking-dangerous
10 https://www.bbc.com/news/articles/cpdlpj9zn9go
11 https://medium.com/@ossiana.tepfenhart/the-disturbing-korean-deepfake-scandal-that-no-one-is-talkingabout-771edaa33e7c
have been misused. Many rushed to delete personal photos from social media and online databases to prevent manipulation. 12 Discussions on forums revealed the alarming nonchalance of some participants. One college student even remarked, "Men have standards too. We don’t just deepfake any woman," revealing a deeply ingrained culture of misogyny and disregard for the devastating consequences of such acts. Victims of the New Nth Room exploitation have shared experiences of psychological distress, social ostracization and deep emotional trauma. One high school student recounted how her life changed overnight when she discovered her face had been used in an explicit video that circulated among her classmates. "I couldn’t leave my house for weeks. I felt like I had lost control of my own identity," she said.
Another victim, a teacher, found herself the subject of deepfake content created by her students. "I dedicated my career to educating these children, and they violated my dignity in the most horrifying way. " she shared. Many victims have reported experiencing anxiety, depression and even suicidal thoughts as a result of the exploitation. This case also contains elements of CASM, as many of the victims in this case were minors whose images were manipulated into explicit content, raising serious concerns about child exploitation and high demand of such material. But due to many of the perpetrators being minors too, it raises a question on how to navigate through this case.
4. LEGAL AND ETHICAL CHALLENGES
a. Inadequacies in Current Legal Frameworks The rise of AI-driven exploitation poses significant challenges to traditional legal frameworks. Many of the laws that govern cybercrime were developed before the advent of AI and are illequipped to address the unique challenges posed by AI-driven exploitation.
1) Deepfakes:
The legal response to deepfakes has been fragmented and inconsistent across nations. In the United States, there is no federal law specifically addressing deepfakes. Some states, such as
12 https://www.hani.co.kr/arti/society/society_general/1155647.html
California13 and Texas, have introduced legislation to criminalize the creation and distribution of deepfake pornography, but these laws are limited in scope and do not address other forms of deepfake misuse, such as political disinformation. The GDPR14 currently in place within the European Union gives some measure of protection against the misuse of personal data, but it does not give clear guidance about deepfakes. It has recently proposed two more pieces of legislation: the Digital Services Act (DSA)15 and the Digital Markets Act (DMA). While these proposals attempt to regulate the environment of online platforms, they are still in their infancy and do not tackle the issue of deepfakes directly. In India, the Information Technology Act, 200016 does not specifically have provisions for addressing deepfakes.
However, Section 66E of the Act, on violation of privacy, may technically be used in the prosecution process against deepfakes creators, but the legal framework is a huge gap area as there isn't specific legislation.
2) AI Generated Child Sex Abuse Material:
In the United States, the Protect Act of 2003 criminalizes the production, distribution, and possession of CSAM. However, the law does not explicitly address AI-generated content, leaving a gray area in its application. In the United Kingdom, the Protection of Children Act
1978 and the Criminal Justice and Immigration Act 2008 also criminalize CSAM but have no
specific provisions for AI-generated material. In Australia, the Criminal Code Act 1995 criminalizes possession and distribution of CSAM. However, like others, it is not addressed to AI-generated content. This legal loophole allows perpetrators the scope for creating and distributing synthetic CSAM with some amount of impunity.
3) Sextortion:
In the United States, sextortion is prosecuted under federal laws such as the Extortion Act and state-level statutes. However, these laws do not specifically address the use of AI to create fake
13 https://leginfo.legislature.ca.gov/faces/billTextClient.xhtml?bill_id=201920200AB602
14 https://gdpr-info.eu/
15 https://digital-strategy.ec.europa.eu/en/policies/digital-services-act-package
16 https://www.meity.gov.in/content/information-technology-act-2000
content, making it difficult to prosecute perpetrators who use AI-powered tools. In the European Union, the Directive on Combating the Sexual Abuse and Sexual Exploitation of Children provides a framework for addressing sextortion, but it does not specifically cover AI-generated content. Similarly, in India, the Information Technology Act, 2000, does not explicitly address sextortion, leaving victims with limited legal recourse.
4) AI Enabled Human Trafficking:
In the United States, the Trafficking Victims Protection Act (TVPA) provides a legal framework for combating human trafficking, but it does not specifically address the use of AI. Similarly, in the European Union, the Directive on Preventing and Combating Trafficking in Human Beings does not explicitly cover AI-enabled trafficking. In India, the Immoral Traffic (Prevention) Act, 1956, and the Information Technology Act, 2000, provide some protections against human trafficking, but they do not address the use of AI. This legal gap allows traffickers to exploit AI technologies with relative impunity. b. Ethical Dilemmas Ethical questions arise about the limits of AI development and the responsibility of developers and corporations in preventing misuse. The debate often centers on whether the benefits of AI outweigh the harm caused by its exploitation.
On one hand, AI promises transformative benefits, driving progress in fields ranging from healthcare to transportation. On the other hand, its rapid development raises significant concerns about misuse and potential harm. AI can revolutionize industries, enhance productivity, and improve quality of life through novel applications. But it also may be used to create realistic and manipulative content that contribute to sexual exploitation or abuse. The ethical dilemma which arises here is- How can society foster the benefits of AI innovation while implementing effective measures to prevent its misuse in sexual exploitative contexts? 17 Establishing legal and ethical guidelines is crucial to prevent AI from being exploited to commit sexual crimes. However, over-regulation may stifle creative and beneficial uses of AI,
17 https://www.geeksforgeeks.org/ethical-considerations-in-ai-development/
while under-regulation can leave gaps that bad actors may exploit. The ethical dilemma which arises here is- Determining the right balance between freedom for technological exploration and ensuring stringent safeguards to protect vulnerable individuals. It is ethically challenging to assign accountability when AI systems are used for sexual exploitation—whether to the developers, users, or the technology itself.18 Greater transparency can help build trust and enable better oversight of AI applications. Thus, establishing clear lines of accountability and ensuring that AI systems are designed with transparency in mind, are the ways AI development can be pursued ethically.19
5. STEPS TAKEN TO MITIGATE SEXUALLY EXPLOITATIVE AI CRIMES
A. Legal Approaches
1. Updating and Amending Existing Legislation:
Many countries are updating their existing legal frameworks to include provisions that address the misuse of digital technologies and AI. For example, sexual exploitation laws and cybercrime statutes are being amended to cover non-consensual content, deepfakes, and other AI-enabled abuses.
2. Artificial Intelligence Act (EU):
The European Union is at the forefront of drafting comprehensive legislation to regulate AI. The proposed Artificial Intelligence Act aims to classify AI systems based on risk, with highrisk systems (including those that could be used to create sexually exploitative material) subject to strict obligations and oversight.
3. Strengthening Cybercrime Conventions and International Cooperation:
The Council of Europe’s Budapest Convention remains a cornerstone for international cooperation against cybercrime. Ongoing discussions aim to update the Convention to explicitly include crimes enabled by modern technologies like AI, ensuring that nations have a 18 https://aiforsocialgood.ca/blog/unveiling-the-ethical-dilemma-in-the-realm-of-artificial-intelligence
19 https://fastercapital.com/startup-topic/Ethical-Considerations-in-AI-Development.html
unified approach in combating these offenses.
4. International Collaboration:
Organizations such as Interpol and Europol facilitate cross-border cooperation among law enforcement agencies. These collaborations are critical for tracking down perpetrators who use AI to produce and disseminate sexually exploitative material.
5. Targeted Legislation:
Some jurisdictions have introduced laws specifically criminalizing the creation and distribution of deepfakes or AI-generated content that constitutes sexual exploitation. Such laws are designed to hold perpetrators accountable even when traditional definitions of “physical” sexual exploitation do not apply.
B. Technological Countermeasures
1. Deepfake Detection
Technological solutions are being developed to detect and counter deepfake content. For example, Facebook and Microsoft have partnered with academic institutions to create the Deepfake Detection Challenge20, which aims to develop AI tools to detect deepfakes.
2. AI-Powered CSAM Detection
AI-powered tools are being used to detect and remove CSAM from online platforms. For example, Google has developed Content Safety API, which uses AI to identify and remove CSAM from its platforms.21
3. Sextortion Prevention
AI-powered tools are also being used to detect and prevent sextortion. For example, Thorn, a non-profit organization, has developed Spotlight, which an AI tool that helps law enforcement
20 https://www.kaggle.com/competitions/deepfake-detection-challenge
21 https://protectingchildren.google/#introduction
identify and rescue victims of sextortion.22
4. AI-Enabled Human Trafficking Detection
AI-powered tools are being used to detect and disrupt human trafficking networks. For example, TraffickCam, an app developed by Exchange Initiative, uses AI to identify hotels where trafficking victims are being held. In 2023, the app was updated to include new features for detecting AI-generated fake identities and documents used by traffickers.23
5. Project Arachnid
Project Arachnid, developed by the Canadian Centre for Child Protection (C3P), is an AIpowered tool that scans the internet for CSAM and issues takedown notices to hosting providers. In 2023, the project expanded its capabilities to include AI-generated CSAM detection.24
6. DeepTrust Alliance's Deepfake Detection Tools
The DeepTrust Alliance, a coalition of tech companies and researchers, launched a suite of deepfake detection tools in 2023. These tools use machine learning algorithms to analyze video and audio content for signs of manipulation. The alliance also provides a public database of known deepfake content to help researchers and law enforcement agencies stay ahead of emerging threats.25
C. Social Interventions
1) Tech Coalition's "Stop Sextortion" Initiative
The Tech Coalition, a group of leading tech companies including Google, Facebook, and Microsoft, launched the "Stop Sextortion" initiative in 2023. This initiative focuses on developing AI tools to detect and prevent sextortion, as well as providing resources for victims and raising public awareness26. 22 https://www.thorn.org/blog/a-new-chapter-for-spotlight-and-thorns-continued-commitment-to-child-safety/
23 https://traffickcam.com/
24 https://projectarachnid.ca/en/
25 https://www.deeptrustalliance.org/
26 https://about.meta.com/actions/safety/topics/bullying-harassment/stop-sextortion
2) UNODC's Blue Heart Campaign
The United Nations Office on Drugs and Crime (UNODC) expanded its Blue Heart Campaign in 2023 to include a focus on AI-enabled human trafficking. The campaign aims to raise awareness about the role of technology in human trafficking and promote international cooperation to combat this crime.27
3) India's "Cyber Safe Women" Initiative
In 2023, the Indian government launched the "Cyber Safe Women" initiative, which focuses on educating women and girls about online safety, including the risks of deepfakes, sextortion, and AI-enabled exploitation. The initiative includes workshops, helplines, and partnerships with NGOs to provide support to victims.
6. THE WAY AHEAD: HARMONIZING TECHNOLOGY, LAW, AND SOCIETY
The fight against AI-driven exploitation requires a coordinated response that harmonizes law, technology, and social action. On the legal front, there is a need for stronger and more comprehensive laws that specifically address the unique challenges posed by AI-driven exploitation. These laws should criminalize the creation and distribution of non-consensual deepfake content, as well as the use of AI-powered tools to coerce and control victims of sextortion and human trafficking. Additionally, there is a need for greater international cooperation to ensure that perpetrators of AI-driven exploitation can be tracked down and prosecuted, regardless of where they operate. On the technological front, there is a need for the development of tools and technologies that can detect and counter AI-driven exploitation.
For example, AI-powered tools could be used to detect and remove deepfake content from online platforms, or to identify and disrupt AIpowered sextortion and human trafficking networks. Technology companies also have a responsibility to ensure that their platforms are not being used to facilitate AI-driven exploitation, and to take proactive steps to prevent such abuses. Finally, social action is needed to raise awareness of the dangers of AI-driven exploitation and to promote a culture of respect and dignity. This includes educating the public about the risks
27 https://www.unodc.org/unodc/en/blueheart/
of deepfake pornography, sextortion, and human trafficking, as well as promoting ethical standards for the development and use of AI technologies.
7. CONCLUSION
AI has the potential to redefine industries and improve lives, but its misuse poses significant risks to individuals and society. Stronger and more comprehensive laws are needed to criminalize the creation and distribution of non-consensual deepfake content, as well as the use of AI-powered tools to coerce and control victims of sextortion and human trafficking. Technology companies are responsible for ensuring that their platforms are not being used to facilitate exploitation, and for developing and using AI-powered tools that can detect and counter harmful content. Civil society organizations also have an important role to play in raising awareness of the dangers of AI-driven exploitation and advocating for stronger legal protections for victims.
By working together, law enforcement, technology companies, and civil society can build a safer digital future, one where technology empowers rather than exploits.
REFERENCES
Primary Sources
1. California Assembly Bill 602 (AB 602), 2019-2020 Legislative Session, California
Legislature. Available at: https://leginfo.legislature.ca.gov.
2. General Data Protection Regulation (GDPR), Regulation (EU) 2016/679, European
Parliament and Council. Available at: https://gdpr-info.eu. 3. Digital Services Act (DSA), European Commission. Available at: https://digitalstrategy.ec.europa.eu.
4. Information Technology Act, 2000, Ministry of Electronics and Information
Technology, Government of India. Available at: https://www.meity.gov.in.
5. Protect
Act of 2003, 108th Congress, United States. Available at: https://www.congress.gov.
6. Protection
of Children Act 1978, United Kingdom. Available at: https://www.legislation.gov.uk.
7. Criminal
Code Act 1995, Australian Government. Available at: https://www.legislation.gov.au.
8. Trafficking Victims Protection Act (TVPA), United States Department of State.
Available at: https://www.state.gov.
9. Directive on Preventing and Combating Trafficking in Human Beings, Directive
2011/36/EU, European Parliament and Council. Available at: https://eur-lex.europa.eu. Secondary Sources
1. Internet
Watch Foundation (IWF), Annual Report
2021 (2021).
Available at: https://www.iwf.org.uk.
2. National Center for Missing and Exploited Children (NCMEC), Sextortion: The
Growing Threat to Minors (2020). Available at: https://www.missingkids.org.
3. United Nations Office on Drugs and Crime (UNODC), Global Report on Trafficking in
Persons (2021). Available at: https://www.unodc.org.
4. Canadian Centre for Child Protection (C3P), Project Arachnid (2023). Available
at: https://www.protectchildren.ca. News Articles and Reports
1. ‘AI-Generated CSAM: A Growing Threat to Child Safety’, BBC News (15 March
2023). Available at: https://www.bbc.com.
2. ‘The Rise of AI in Human Trafficking: How Technology is Being Exploited’, The
Guardian (20 May 2023). Available at: https://www.theguardian.com.
3. https://www.hrw.org/news/2024/07/01/using-ai-fight-trafficking-dangerous
4. https://www.bbc.com/news/articles/cpdlpj9zn9go
5. https://medium.com/@ossiana.tepfenhart/the-disturbing-korean-deepfake-scandalthat-no-one-is-talking-about-771edaa33e7c
6. https://www.hani.co.kr/arti/society/society_general/1155647.html
Websites
1. Deepfake Detection Challenge: https://deepfakedetectionchallenge.ai.
2. Thorn’s Spotlight Tool: https://www.thorn.org.
3. TraffickCam App: https://www.traffickcam.com.
4. Project Arachnid: https://www.protectchildren.ca.
5. DeepTrust Alliance: https://www.deeptrustalliance.org.
6. Tech Coalition’s Stop Sextortion Initiative: https://www.techcoalition.org.
7. UNODC Blue Heart Campaign: https://www.unodc.org.
8. India’s Cyber Safe Women Initiative: https://www.meity.gov.in.
9. https://about.meta.com/actions/safety/topics/bullying-harassment/stop-sextortion
10. https://www.dazeddigital.com/science-tech/article/55926/1/inside-the-disturbing-riseof-deepfake-porn 11. https://texta.ai/blog/ai-technology/the-rise-of-deepfake-porn-exploring-thecontroversies-and-consequences 12. https://www.dazeddigital.com/science-tech/article/55926/1/inside-the-disturbing-riseof-deepfake-porn
13. https://www.geeksforgeeks.org/ethical-considerations-in-ai-development/
14. https://aiforsocialgood.ca/blog/unveiling-the-ethical-dilemma-in-the-realm-ofartificial-intelligence
15. https://fastercapital.com/startup-topic/Ethical-Considerations-in-AI-
Development.html 16. https://aiforsocialgood.ca/blog/unveiling-the-ethical-dilemma-in-the-realm-ofartificial-intelligence
17. https://legal-wires.com/lex-o-pedia/understanding-sextortion-forms-elements-andlegal-context/
18. https://www.nationalcrimeagency.gov.uk/news/nca-issues-urgent-warning-aboutsextortion\
19. https://www.cameraforensics.com/blog/2024/11/18/a-guide-to-ai-generated-csam-forinvestigators-of-online-exploitation/
20. https://ssslearning.co.uk/safeguarding-articles/AI-generated-child-abuse/
21. https://www.missingkids.org/blog/2024/generative-ai-csam-is-csam
22. https://enoughabuse.org/get-vocal/laws-by-state/state-laws-criminalizing-aigenerated-or-computer-edited-child-sexual-abuse-material-csam/
23. https://digital-strategy.ec.europa.eu/en/policies/european-approach-artificialintelligence