Artificial Intelligence, Deepfakes and Sexual Harassment at the Workplace: Reimagining the Indian POSH Act in the Digital Age
Syed Abdul Khader Jeelani | MBA., LLM (PhD)
ABSTRACT
Artificial Intelligence (AI) has transformed communication, business processes, and workplace interactions in a significant manner. However, along with its immense benefits, AI has inter alia generated new forms of violence, including gender-based violence, by using deepfake pornography, synthetic sexual imagery, voice cloning, and AI-enabled online harassment, which is not ruled out in the modern era and the corporate world. These technological developments have blurred the boundaries between physical and virtual workplaces, thereby posing significant challenges to enforce and implement the Sexual Harassment of Women at Workplace (Prevention, Prohibition and Redressal) Act, 2013 (POSH Act). While the POSH Act was enacted to provide a safe and dignified workplace for women, it was drafted before the emergence of generative AI and does not expressly recognise AI-assisted sexual harassment. This article examines whether AI-generated sexual misconduct falls within the existing legal framework under the POSH Act, analyses the risks posed by deepfake technologies, identifies deficiencies in the present legal regime, and proposes legislative and institutional reforms to ensure that Indian workplaces remain safe in the era of artificial intelligence.
Keywords: Artificial Intelligence, Deepfake, POSH Act, Workplace Sexual Harassment, Digital Evidence, Online Harassment, India.
Introduction
The rapid advancement of Artificial Intelligence (AI), particularly generative AI, has revolutionised digital communication. AI applications can generate realistic images, videos, audio recordings, and textual content that closely resemble real individuals. One of the most concerning manifestations of this technology is the creation of “deepfakes”—synthetically generated or manipulated media that falsely depicts a person engaging in acts or speech that never occurred. While AI has significantly enhanced productivity and innovation, it has simultaneously created unprecedented risks for workplace dignity, privacy, and gender equality. The celebrities were vulnerable to malicious exploitation and attack by AI with realistic images, video and voice cloning. These technologies increase the risk of reputational injury and invasion of privacy and may not only cause Economic Loss but also severe psychological trauma to the women employees.Â
Sexual harassment is no longer confined to physical spaces. Workplace harassment increasingly occurs through emails, messaging platforms, video conferencing applications, social media, and AI-generated digital content, which get widely spread in Social Media like Instagram, Facebook, WhatsApp, Telegram, LinkedIn and other Platforms. Female employees may become victims of fabricated sexually explicit images, manipulated videos, cloned voices, or AI-generated defamatory content circulated among colleagues. Such acts can destroy professional reputation, create hostile work environments, and inflict severe psychological trauma even without any physical contact.
Some instances of US Courts are reproduced for understanding as how courts are stringent abroad. A California appellate court recently affirmed a jury verdict awarding $4 million to a police captain who was subjected to a hostile work environment after a sexually explicit, AI-generated image resembling her was widely circulated in the workplace, holding that the dissemination of such fabricated content constituted unlawful harassment under California law.In another case, a Washington State trooper filed suit alleging that a supervisor used AI to create and circulate a deepfake video of him intimately kissing a coworker; the officer is suing his employer for discrimination, retaliation, and invasion of privacy. These high-profile incidents highlight a disturbing trend: AI-generated content—especially deepfakes—is emerging as a powerful new form of workplace harassment.
The POSH Act, enacted in 2013 following the principles laid down in Vishaka v. State of Rajasthan (1997), primarily addresses conventional forms of workplace sexual harassment. However, the legislation does not expressly address AI-generated abuse, leaving considerable ambiguity regarding employer liability, investigation procedures, and digital evidence. This emerging gap necessitates a re-examination of the legal framework.
AI-Driven Sexual Harassment and Deepfake Technology
Deepfake technology employs machine learning algorithms to create highly convincing fabricated images, videos, and audio recordings. Although the technology has legitimate applications in education, entertainment, and accessibility, its misuse has become increasingly common.
Within workplaces, AI-enabled sexual harassment may include:
Creation and circulation of non-consensual sexually explicit deepfake images of employees;
AI-generated pornographic videos using an employee’s face;
Voice cloning to simulate sexually inappropriate conversations;
AI-generated sexually explicit emails or messages impersonating employees;
Morphing photographs into obscene content;
Synthetic manipulation during virtual meetings or on online platforms.
Unlike traditional sexual harassment, AI-generated abuse may occur anonymously, across jurisdictions, and without any physical interaction. The content can spread rapidly across digital platforms, making removal extremely difficult and the deduction of the source also difficult, thereby causing irreversible reputational damage. The psychological consequences include anxiety, depression, social withdrawal, loss of employment opportunities, reputational injury, emotional distress, and reduced workplace participation. Victims frequently experience secondary victimisation when employers lack adequate technological expertise to investigate digital complaints.
Applicability of the POSH Act to AI-Based Harassment
The POSH Act defines sexual harassment under Section 2(n) to include unwelcome physical, verbal, or non-verbal conduct of a sexual nature. Section 3 further guarantees every woman the right to a safe workplace free from sexual harassment.
Although the Act does not specifically mention AI-generated content, several forms of deepfake abuse may arguably fall within existing provisions because they constitute unwelcome sexually coloured conduct, humiliation, intimidation, or creation of a hostile work environment.
For example:
circulation of AI-generated obscene images of a colleague;
creation of sexually explicit digital content using an employee’s identity;
dissemination of manipulated videos intended to embarrass or intimidate women employees;
Use of AI chatbots or cloned voices for sexually offensive communications.
Similarly, the broad definition of “workplace” under the POSH Act extends beyond physical office premises and includes virtual interactions connected with employment. Therefore, incidents occurring through Microsoft Teams, Zoom, WhatsApp groups, workplace emails, or enterprise communication platforms may also attract the provisions of the Act. So the definition to be further extended to include all social Media Platforms and Networking Platforms.
Internal Committees (ICs) are therefore likely to possess jurisdiction and also awareness/knowledge where AI-generated conduct creates a hostile work environment or amounts to sexual harassment.
Interface with Other Indian Laws
1. While the POSH Act provides civil workplace remedies, several other Indian laws may become applicable depending upon the facts.
2. The Information Technology Act, 2000 criminalises the publication or transmission of obscene and sexually explicit electronic material under Sections 67 and 67A.
3. The Bharatiya Nyaya Sanhita, 2023 contains provisions relating to obscenity, voyeurism, insult to the modesty of women, criminal intimidation, identity-related offences, and publication of sexually explicit material that may apply to deepfake misuse.
4. Additionally, the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021 require intermediaries to remove unlawful content upon receiving a valid legal notice.
However, these statutes were not drafted with generative AI in mind and do not comprehensively regulate synthetic media or impose clear obligations regarding AI-generated sexual abuse.
Deficiencies in the Existing Legal Framework
Despite the broad language of the POSH Act, significant shortcomings remain.
a. Firstly, there is no statutory definition of AI-generated harassment, deepfakes, synthetic media, or digital impersonation. Consequently, Internal Committees often face uncertainty while determining jurisdiction and require expert knowledge of technology and law.
b. Secondly, the Act lacks detailed mechanisms and procedures regarding the collection, preservation, authentication, and admissibility of digital evidence such as manipulated videos, metadata, cloud-based communications, and AI-generated files. There are no guidelines set so far.
c. Thirdly, the employers generally do not formulate AI-specific workplace policies addressing misuse of generative AI tools. So there is a need to frame a policy for Generative AI tools and suggest remedies and procedures.
d. Fourthly, since committee members frequently lack technical expertise necessary to distinguish genuine digital evidence from AI-generated manipulations. Frequent training and domain knowledge to be updated
e. Fifth, there is inadequate coordination between Internal Committees, cybercrime authorities, digital forensic laboratories, and law enforcement agencies. A brief procedural guideline integrating all the authorities and organisations should be aligned, and a comprehensive mechanism to be framed and followed to give an effective remedy.
Finally, the rapid dissemination of deepfake content frequently causes irreparable reputational damage before legal remedies become effective. Hence, preventive and detective measures, along with punitive damages and deterrent punishment to be imposed so that the mischief is not repeated by others.
Risks and Emerging Challenges
AI-enabled sexual harassment presents several unique risks that distinguish it from conventional workplace misconduct.
1. The anonymity of offenders significantly complicates investigation and accountability. Deepfake content can be created remotely by co-workers, contractors, third parties, or even former employees. Thereby causing detection to be a significant challenge.
2. Cross-border digital platforms further complicate enforcement because offending content may be hosted outside India, and though the John Doe orders passed by Indian courts may not have an impact unless the delinquent is arrested and the content is removed permanently.
3. The speed of viral dissemination makes removal difficult despite subsequent legal intervention.
4. Victims frequently suffer professional stigma despite being innocent because fabricated content appears highly realistic.
5. AI also lowers the cost and technical barriers associated with creating sexually explicit fake content, making such abuse increasingly accessible without any huge effort or planning.
These challenges demand both technological preparedness and legal modernisation.
In cases involving AI-generated deepfakes or synthetic media, Indian courts are likely to grant a combination of:
urgent interim and permanent injunctions;
take-down and blocking orders directed at intermediaries;
protection of the individual’s voice, facial features, and other personality attributes;
damages or an account of profits where commercial exploitation is established; and
Ancillary orders to prevent further dissemination of infringing content.
These remedies collectively aim to safeguard an individual’s identity, reputation, privacy, and dignity while adapting established principles of personality or any women Employees rights to emerging AI-driven forms of digital impersonation.
Recommendations for Reform
The emergence of AI-enabled workplace harassment requires a multidimensional regulatory response.
a. First, the POSH Act should be amended to expressly recognize AI-generated sexual harassment, deepfakes, voice cloning, synthetic imagery, digital impersonation, and technologically facilitated sexual misconduct.
b. Second, employers should be statutorily required to adopt AI governance policies that prohibit misuse of generative AI against employees.
c. Third, Internal Committees should receive mandatory training in digital evidence preservation, cyber investigations, and AI-assisted harassment.
d. Fourth, organisations should establish rapid-response mechanisms for immediate reporting, preservation, takedown requests, and victim assistance where deepfake content is involved.
e. Fifth, collaboration between Internal Committees, CERT-In, cybercrime police stations, and accredited digital forensic experts should be institutionalised.
f. Sixth, confidential psychological counselling and reputation-restoration mechanisms should be incorporated into workplace grievance systems.
g. Finally, Parliament should consider introducing comprehensive legislation regulating AI-generated synthetic media, including civil liability, criminal sanctions, mandatory watermarking of AI-generated content, platform accountability, and victim compensation.
Conclusion
Artificial Intelligence has fundamentally altered the nature of workplace interactions. Deepfake technology and AI-generated sexual misconduct present new challenges that were largely unimaginable when the POSH Act was enacted in 2013. Although the Act’s broad language permits an interpretation that includes certain forms of AI-assisted harassment, the absence of explicit statutory recognition creates uncertainty in implementation and enforcement.
The law must evolve alongside technological innovation. Legislative amendments, institutional capacity building, AI governance frameworks, digital forensic expertise, and stronger collaboration between employers and cyber authorities are essential to preserve workplace dignity. A proactive legal framework will ensure that technological advancement serves the cause of equality rather than becoming a tool for gender-based exploitation. The future ofworkplace safety depends upon integrating traditional principles of human dignity with modern technological realities.
References (APA 7th Edition)
Cath, C. (2018). Governing artificial intelligence: Ethical, legal and technical opportunities and challenges. Philosophical Transactions of the Royal Society A, 376(2133).
Government of India. (2000). Information Technology Act, 2000.
Deepfakes in the Workplace: The Emerging Legal Risks of AI-Driven Harassment, published in Wolters Kluwer
Government of India. (2013). The Sexual Harassment of Women at Workplace (Prevention, Prohibition and Redressal) Act, 2013.
Government of India. (2021). Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021.
Government of India. (2023). Bharatiya Nyaya Sanhita, 2023.
Kietzmann, J., Lee, L., McCarthy, I., & Kietzmann, T. (2020). Deepfakes: Trick or treat? Business Horizons, 63(2), 135–146.
Supreme Court of India. (1997). Vishaka v. State of Rajasthan, (1997) 6 SCC 241.
UNESCO. (2021). Recommendation on the Ethics of Artificial Intelligence.
West, S. M., Whittaker, M., & Crawford, K
. (2019). Discriminating Systems: Gender, Race and Power in AI. AI Now Institute.
