Artificial Intelligence: A Powerful Tool or the Next Challenge to Privacy, Identity, and Human Dignity?

AI Is Changing Lives — For Better and for Worse

Artificial Intelligence (AI) is transforming how we live, work, communicate, and make decisions. In India, government agencies, banks, hospitals, schools, and businesses are racing to adopt AI-driven tools. For ordinary people, AI powers language translation apps, personalised recommendations, and digital payments. It helps doctors spot diseases. It aids farmers with crop management. It even assists lawyers with research. UNESCO’s AI Ethics Recommendation underscores that these technologies must “respect human rights and human dignity.” Yet great power brings great responsibility.

At the same time, AI is creating new kinds of risk. Criminals can now generate eerily convincing fake voices. They can manipulate photos and videos to create so-called “deepfakes.” They can craft highly personalised phishing attacks. Even familiar voices and faces no longer prove identity reliably. This technological revolution is testing our trust in an unprecedented way. We must ask: are our laws, institutions, and public awareness keeping up?

The Benefits: AI Empowering People and Institutions

AI has already improved productivity and access in almost every sector. Many farmers use smartphone AI apps to analyse soil and monitor crops. Doctors use AI algorithms to detect early signs of disease. Students get personalised tutoring. Judges and attorneys can quickly search vast legal databases using AI tools.

Even courtrooms are experimenting with AI. Software like SUVAS translates judgments into regional languages. SUPACE summarises case files for busy judges. These innovations show that AI can empower people and institutions alike.

The Dark Side: How Criminals Exploit AI

Alongside these benefits, AI is changing the game for criminals. Scammers now use AI to impersonate anyone. In one Indian case, fraudsters cloned a family member’s voice via AI. They tricked a 72-year-old woman into transferring ₹1.97 lakh to them. It started with a WhatsApp plea from her “sister-in-law.” When the woman called back, the voice sounded exactly like her relative. The scammers convinced her using urgency and fear, and the money vanished before anyone realised what had happened.

The Indian Express reports that 83% of Indian victims of AI voice scams lost money, and nearly half lost more than ₹50,000. Moreover, 69% of Indians can’t distinguish a real voice from an AI clone. It isn’t just the elderly who are vulnerable: about 47% of Indian adults personally know someone who fell prey to an AI-based scam. Scammers exploit human trust and urgency across WhatsApp, SMS, and phone calls, layering AI tricks like voice cloning, chatbots, and deepfake calls onto familiar scam formats.

The result: trust itself is under attack. We can no longer take what we hear or see online at face value. This dual nature of AI — immense power paired with real peril — sets the context for everything that follows. We must stay clear-eyed about both sides. We cannot let fear freeze innovation, but we must also confront these new dangers responsibly.

The Growing Threat to Privacy and Identity

In India, our digital-age identity extends far beyond a passport or Aadhaar card. It now includes our online footprint: face and voice data, email addresses, social media profiles, photographs, financial details, browsing history, and more. Indian law declared privacy a fundamental right in Justice K.S. Puttaswamy v. Union of India (2017). The Supreme Court held that privacy is intrinsic to the dignity and liberty that Article 21 of the Constitution guarantees. In other words, privacy isn’t an optional luxury — it’s a core constitutional value.

Protecting privacy matters more than ever in the AI era. UNESCO’s AI Ethics Recommendation states that privacy must be protected throughout the AI lifecycle, from data collection to processing to storage. Indian law now reflects this principle directly. Parliament enacted the Digital Personal Data Protection Act, 2023 on 11 August 2023. The DPDP Act builds a framework for safeguarding personal data. It requires organisations to process data fairly and transparently, and it grants individuals the right to know, correct, or delete their data.

India’s Information Technology Act, 2000 (amended, and read with its rules) also provides penal provisions for many cyber offences. Section 66C criminalises identity theft. Section 66D covers cheating by personation. Section 66E penalises capturing someone’s image without consent. The Act also prohibits publishing or transmitting obscene material, including child pornography (Sections 67, 67A, 67B). These laws give law enforcement real tools to prosecute impersonation or harassment — though they predate today’s AI tools.

Authorities updated the rules for internet intermediaries in 2021 and amended them again in 2026 to explicitly address synthetic media. In early 2026, the government notified amendments requiring platforms to take “appropriate technical measures” to prevent unlawful AI-generated content, including deepfakes that are obscene or harmful to children. Platforms must now clearly label allowed synthetic content, make it traceable, and set up faster takedown processes. The amended rules demand removal of illicit content, such as non-consensual intimate imagery, within three hours of notice.

Finally, a new criminal code — the Bharatiya Nyaya Sanhita, 2023 (enacted December 2023) — has replaced the old IPC. It consolidates offences including cheating, defamation, identity crimes, and cybercrime. The BNS uses different section numbers, but it still outlaws personation, fraud, and wrongful disclosure of private information. Together, these evolving laws show that India is moving to cover digital and AI-enabled harms under the law.

AI-Generated Deepfakes: A New Form of Digital Harm

What Deepfakes Are and Why They Matter

One of the most alarming developments is the rise of AI-generated deepfakes. A “deepfake” is any synthetic video or audio that convincingly portrays a real person doing or saying something they never did. Advances in generative AI mean anyone with internet access can create a realistic fake video in minutes. The implications for privacy and dignity run deep. Criminals and abusers use such content in several damaging ways:

• Reputation damage — a fake video of a public figure or professional can ruin careers or sway public opinion.
• Misinformation — a deepfake of a leader making inflammatory statements could incite unrest; a fake emergency video of a family member can spark panic.
• Financial fraud — an AI-cloned voice of a relative can trick someone into wiring money under false pretences.
• Blackmail and extortion — cybercriminals fabricate intimate videos using victims’ photos, then demand money to keep them private.
• Harassment and humiliation — non-consensual “deepfake porn” often circulates online, shaming victims who never consented or even participated.

Real Cases and the Scale of the Problem

Criminals can weaponise such content against anyone, famous or ordinary. A November 2023 case in India shows the danger clearly. Cybercriminals created an AI-synthesised video of a real senior police officer, complete with a matching voice, to extort a 76-year-old man. The victim believed the fake officer would charge him with a crime, so he paid over ₹74,000 out of fear. He even contemplated suicide before his family intervened. Reports named this one of India’s first deepfake extortion cases.

Research shows the scale of this problem is staggering. Globally and in India, nearly all deepfake abuse targets women. A 2026 report by tech researchers found that 93% of “explicit” deepfake content in India targets women, and the volume of such content has surged by nearly 900% in recent years. Worldwide, about 98% of deepfake pornography targets women. A separate study found that 96% of examined deepfake video clips contained non-consensual intimate content, and 100% of top deepfake porn sites featured only women. As one analysis explains, non-consensual intimate deepfakes cause the same well-established harms as non-synthetic image-based sexual abuse. Victims endure severe stress, anxiety, depression, and low self-esteem — just like victims of real abuse.

For ordinary people, a viral deepfake can shatter a life. Fake images circulated on social media can cause shame at school or work. Imposters on dating sites can use deepfake photos to lure unsuspecting people. Children often share photos online without realising the risks, which makes them targets for deepfake bullying. In every case, an AI-related crime translates into very human suffering. Victims report severe embarrassment, mental trauma, and a feeling that their privacy and personhood have been violated. Deepfakes aren’t just a technical issue — they’re a human rights issue.

The Human Cost of AI Misuse

Behind every statistic stands a real person. Consider a few examples of how AI-enabled crimes affect individuals:

• Personal reputation — a small-business owner discovers a fake video that appears to show him making a bribe. Customers share the clip, and he spends months proving it was fabricated.
• Emotional trauma — a college student withdraws after an AI-generated explicit image of her circulates among peers. She struggles with anxiety and fear of shame.
• Family betrayal — a grandmother in Haryana loses her savings when criminals use an AI voice clone of her grandson to ask for money convincingly. She never recovers the funds.
• Social harassment — a schoolgirl in Bangalore faces online bullying after strangers create a fake social media profile of her, posting abusive comments and edited photos.
• Blackmail — a young woman in Mumbai finds a fabricated intimate video of herself online. Scared of the damage to her reputation, she quietly pays off the extortion demand rather than approach the police.

These stories show that AI-fuelled crimes deeply affect people’s mental health, relationships, careers, and dignity. Victims of non-consensual deepfakes suffer violations of privacy, dignity, sexual expression, and mental well-being that closely mirror the harms of real image-based abuse. When criminals target women in public life, it creates an additional chilling effect: it discourages other women from entering politics or leadership roles for fear of similar attacks. These harms aren’t new in form, but they demand new solutions.

AI and Online Financial Frauds

Common AI-Driven Scam Tactics

Cyber fraud has become alarmingly sophisticated because of AI. Criminals use AI to craft convincing lures and lower the bar for attacks. Deepfake technology now enables what experts call “vishing,” or deepfake voice phishing. Scammers scrape a target’s social media for voice samples or personal details. They train a voice-cloning model, then call the victim using a familiar-sounding voice to demand money or information.

Common AI-driven scams include:

• Voice cloning scams — fraudsters lift a person’s voice from videos and clone it to send urgent pleas over WhatsApp or phone calls.
• OTP and phishing scams — criminals use AI to guess or intercept one-time passwords through call-merging or SIM swapping, fooling victims into sharing bank OTPs.
• Digital arrest scams — scammers pose as police or government officers in fake video calls, using AI-generated backgrounds and voices to intimidate victims into transferring money.
• Loan and app scams — malicious apps harvest contacts and photos; fraudsters then use AI to fabricate threats or images to extort victims.
• Investment and romance scams — AI-generated chatbots and videos impersonate attractive dates or investment opportunities to persuade victims to send money.
• Deepfake extortion — attackers create fake intimate videos of victims using stolen photos, then blackmail them for hush money.

Why Awareness Still Matters Most

AI doesn’t power every new scam, but it makes old scams far more effective. As one cybersecurity expert put it, AI “layers onto traditional frauds to make them more believable and harder to detect.” Voice cloning, romance scams, and deepfake extortion pose the biggest threats today.

Governments and companies are developing detection tools in response, but public awareness remains crucial. Indian cyber analysts warn that many victims act in good faith simply because a scam appears genuine. The Hyderabad cybercrime unit advises people to always double-check urgent money requests from family, even when the voice sounds familiar. They recommend a few precautionary steps: place a video call to verify the person, enable security features on your apps, and report suspicious requests immediately.

Women and AI-Enabled Online Abuse

Women faced disproportionate online targeting even before AI existed, and AI has deepened this disparity. As shown above, nearly all deepfake porn targets women. Similarly, many social media smear campaigns and fake profiles try to discredit women in public life. AI makes it trivial to overlay a woman’s face onto lewd images or fake dialogues.

This creates severe real-world harm. Victims of image-based abuse, including deepfakes, often experience high levels of stress, anxiety, depression, low self-esteem, and insecurity. They may face workplace harassment or ostracism, health issues like sleep loss, and social isolation from shame. A political activist targeted by a humiliating deepfake, for example, might withdraw from public work out of fear of backlash. In extreme cases, victims turn to self-harm.

Despite these risks, many incidents go unreported. A March 2026 report found that globally, 62% of deepfake harassment cases involving women never get reported, largely due to stigma. In India, more than one-third of women who experience online abuse take no formal action, and some reduce their online presence out of fear. Alarmingly, about one in three Indian women surveyed didn’t know the laws that protect them from cyber-harassment.

We must recognise that digital violence is real violence. Section 354C of the IPC (retained in the BNS) already punishes voyeurism, and Section 509 punishes words or gestures intended to insult modesty. These laws apply to online harassment and deepfake imagery too. The BNS also explicitly criminalises disclosing a sexual offence victim’s identity (Section 72) to protect victims. However, law enforcement and public awareness must catch up to make these provisions effective against cybercrime.

Children Are the Most Vulnerable

Children and teenagers live in a “digital-first” world. Many share photos, videos, or personal details on social media and apps without fully understanding the long-term risks. AI tools can exploit this innocence. Criminals can scrape a child’s selfie to create a deepfake, or an AI chatbot can lure a young gamer by pretending to be a friendly stranger.

The data confirm that cyber threats to children are climbing. According to India’s National Crime Records Bureau, cybercrimes against minors jumped from 232 cases in 2018 to 1,823 cases in 2022. This includes offences like online sexual abuse, harassment, fake profiles, and bullying. The government has launched several initiatives in response. It set up the Indian Cyber Crime Coordination Centre (I4C) in July 2024, and the National Cyber Crime Reporting Portal (cybercrime.gov.in) now lets anyone file a complaint about online crimes. A toll-free helpline, 1930, offers further support. On the education front, the Ministry of Education’s PRAGYATA Guidelines (2020) and the NCERT curriculum now emphasise safe and responsible technology use for students.

Parents and teachers must stay engaged. They should teach children safe social media habits, such as using privacy settings and thinking before sharing. They should also explain the risks of talking to strangers online, the danger of clicking unknown links, and how to recognise AI-driven scams or inappropriate content. Schools can help too, by setting up Cyber Safety Clubs and teaching digital hygiene. Digital literacy is as vital as reading and math in the 21st century. A child who knows not to share personal details with strangers, or to question a strange request, stays far safer online.

AI in the Legal Profession

AI is changing the legal world too. Indian courts and lawyers are beginning to use AI tools for routine tasks. Predictive analytics can optimise case scheduling, while natural language processing tools help translate judgments into regional languages. Lawyers use AI-powered legal databases to find precedents faster. Even ChatGPT-like services can draft contracts or summarise cases.

This technological aid helps, but it has limits. Legal advice and judgment require deep human reasoning, ethical consideration, and an understanding of social context — qualities machines cannot yet replicate. The Supreme Court’s pilot AI tools, like SUVAS (translation) and SUPACE (case summarisation), operate with disclaimers and human oversight. As the Chief Justice of India has noted, AI must assist lawyers and judges, not replace them. In short, AI can boost efficiency in legal practice, but the core of justice — fairness, empathy, accountability — remains firmly human.

India’s Legal Framework for AI Harms

India already has several laws that prosecutors can invoke against AI-related crimes, depending on the facts:

• The Information Technology Act, 2000 — sections on hacking, data theft, identity fraud, and abusive content (Sections 43, 66, 66C–66E, 67A–67B) can cover AI-enabled identity theft or obscene content.
• The Digital Personal Data Protection Act, 2023 — India’s new data privacy law requires entities to keep personal data secure and obtain consent for its use. It creates a Data Protection Board for enforcement, and violations, like a breach exposing AI training data, can attract heavy fines.
• The Bharatiya Nyaya Sanhita, 2023 — criminalises cheating (Section 181), defamation (Section 196), and blackmail (Section 200) under new numbering. Prosecutors can pursue extortion via a cloned voice under BNS provisions against fraud and criminal intimidation.
• The IT (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021 (amended 2026) — these rules require platforms to remove unlawful content and block fake accounts and deepfake media. The 2026 amendments explicitly cover AI-generated “malicious synthetic media,” obligating platforms to label or block it.
• Other applicable laws — the Code of Criminal Procedure contains sections on cybercrime investigation (like 78A–78Z), and the new Bharatiya Sakshya Adhiniyam (Evidence Act, 2023) provides for forensic reporting on digital evidence.

The National Cyber Crime Reporting Portal (cybercrime.gov.in) makes it easier for victims to lodge complaints online, backed by the 1930 toll-free helpline. Once someone files a complaint, state and city cyber cells are expected to convert it into an FIR and investigate under the relevant law. However, ground realities suggest many victims still don’t report incidents, due to fear or lack of awareness, which hampers enforcement.

In summary, no AI-specific penal statute exists yet, but the current legal framework broadly covers most abuses. Misuse of one’s digital identity, unauthorised access, fraud, cyber harassment, and publishing obscene material are all offences under Indian law. The real challenge lies in applying these laws to new contexts and updating them as AI evolves.

Landmark Judicial Developments

Several landmark court rulings shape India’s approach to technology, privacy, and free speech:

• Puttaswamy v. Union of India (2017) — the Supreme Court unanimously held that privacy is a fundamental right under the Constitution. This case, originally about Aadhaar, established that personal autonomy and dignity demand protection of private data and personal choices. Its principles apply directly to AI: any digital tracking, profiling, or intrusion using AI must meet strict constitutional tests.
• Shreya Singhal v. Union of India (2015) — this judgment struck down Section 66A of the IT Act, calling it a vague restriction on free speech. It also upheld safe-harbour provisions for intermediaries, but added that platforms can take down online content only under a court order or a lawful government notice. In practice, this means platforms cannot arbitrarily remove user content, including suspected deepfakes, without due process.
• Other digital rights cases — courts have since cited the Puttaswamy ruling in decisions upholding LGBT rights (Navtej Johar) and women’s autonomy (Joseph Shine), reinforcing its broad scope. Courts have also used Sections 67 and 67A to prosecute cases of “revenge porn” and child exploitation online.

Together, these decisions form an important backdrop. They confirm that constitutional rights under Article 21 and free speech under Part III extend into cyberspace. They also signal that any new AI-related laws must pass the constitutional test of reasonableness and proportionality.

How Can We Protect Ourselves?

Technology alone cannot defend us — awareness and good habits matter just as much. Experts recommend these practical steps:

• Use strong, unique passwords for every account, and enable two-factor authentication wherever possible.
• Never share OTPs, PINs, passwords, or sensitive information, even if someone calls or messages urgently claiming to be your bank or a relative. Legitimate institutions never ask for your one-time password. If in doubt, hang up and verify through official channels.
• Verify the identity of anyone requesting money or action. Cross-check unusual requests via a video call or through another family member. Criminals often manufacture urgency to discourage verification, so stay calm and double-check.
• Pause before acting on unusual requests. Scammers create false urgency or fear (“act now or your account will be closed”). Take a moment to breathe, assess the situation, and discuss it with someone you trust.
• Limit the personal information you share online. The fewer details — birthdates, relatives’ names, workplace — that are publicly available, the harder it becomes for AI to impersonate you or tailor a scam. Review your app privacy settings regularly.
• Educate children and family members about online safety. Teach kids not to chat with strangers or share photos widely, and encourage them to speak up if something online feels off.
• Preserve evidence if someone targets you. Take screenshots, save call records, and note suspicious emails or URLs — this documentation helps authorities investigate. Report incidents promptly through the National Cyber Crime Reporting Portal (cybercrime.gov.in) or by calling 1930. Quick reporting improves the odds that law enforcement can act before more damage occurs.

The Way Forward

AI stands as one of humanity’s greatest innovations, holding enormous promise for India to leap ahead in development and services. But its future impact will depend not on the machines themselves, but on the values we choose to apply to them. Progress must go hand in hand with ethics: transparency, accountability, fairness, and respect for human dignity.

• Government and policymakers must continue updating laws and enforcement mechanisms. This means implementing the DPDP Act in letter and spirit, strengthening cyber forensics and law enforcement capacity, and ensuring the new IT Rules (2026) work effectively to remove malicious deepfakes and harassment content.
• Technology companies and platforms play a vital role. They should build safer AI: clear watermarks on AI-generated content, robust deepfake detection tools, and clear mechanisms for victims to flag and remove offensive material. Companies should also audit their AI systems internally for bias or privacy leaks.
• Educational and civil society institutions should promote digital literacy and AI awareness. Schools and colleges can teach online safety and critical media consumption, while digital literacy drives (through NSS or NCC, for example) can reach adults too. As UNESCO recommends, open education and media literacy programmes should foster public understanding of AI, empowering citizens to recognise manipulation and use technology wisely.
• Families and communities need open conversations. Parents should talk to children and elders about online risks without blame. Workplaces should update their training for remote and digital work, covering phishing and deepfake threats.

Above all, every individual should remember that privacy and dignity aren’t just personal choices — they’re shared social values. The sanctity of one’s identity and personhood is at stake online just as it is offline. As citizens, we must actively protect ourselves and others. As a society, we must demand that AI innovation follow ethical principles. If we do, AI can remain a force for progress rather than a tool for exploitation.

Ultimately, the future of Artificial Intelligence will not be defined by machines. It will be defined by the values with which human beings choose to use them.