AI Companions and Child Safety: Regulating Chatbots After California’s Adam’s Law
California Tightens AI Child Safety
- On September 10, 2026, California passed Senate Bill 1119 (“Adam’s Law”) to strengthen safety rules for AI companion chatbots used by children, following concerns over risks to children’s mental well-being.
- The law is named after Adam Raine, a 16-year-old who died by suicide in April 2025; his family has filed a case against OpenAI, which is ongoing.
- India currently has no specific law addressing the psychological risks of AI companion chatbots for children, making California’s law a useful reference.
Key Safeguards Under Adam’s Law
- Adam’s Law includes crisis-detection protocols, parental controls, alerts when children disable safety settings, and independent safety audits.
- Operators must either determine each user’s age or apply child protections to all users. The law also restricts advertising directed at children.
- Several obligations come into force later, with the broader framework starting in 2027.
India’s Existing Legal Framework and Its Gaps
| Law (India) | Coverage | Gap |
|---|---|---|
| IT Act, 2000 | The main framework for cybercrime and intermediaries. | Not designed for psychological harm by design |
| DPDP Act, 2023 | Requires verifiable parental consent for processing children’s data. | Silent on emotional dependence and manipulation |
| POCSO Act, 2012 | Addresses sexual offences against children, including online. | Does not cover non-sexual psychological harm |
Multi-Dimensional Concerns
Polity
Article 21 and Article 39(f) place a duty on the State to protect children’s dignity and development, and this duty now extends to digital spaces.
Governance
Existing cybercrime machinery is built for conventional offences, so investigators struggle to examine algorithmic behaviour or use chatbot logs as evidence.
Social Sector
Children turn to AI companions partly because of limited access to mental health support, so banning them without alternatives would leave this need unaddressed.
Science and Technology
Harm can arise from product design, such as engagement-maximising and validating responses, so accountability must cover design and not only content.
Ethics
The main stakeholders are children, parents, developers and regulators. The core dilemma is between a child’s privacy and autonomy and a parent’s right to oversee.
Why AI Child Safety Matters
- Child protection: Early rules can prevent harm before it spreads widely among young users.
- Platform accountability: Clear duties shift responsibility from families to the companies that design these products.
- Regulatory leadership: A balanced framework would position India as a norm-setter on AI governance in the Global South.
Key Regulatory Challenges
Legal Gap
No Indian law deals with emotional dependence or manipulation by chatbots.
Enforcement Capacity
Agencies lack the technical skills to audit algorithms and treat chat logs as evidence.
Privacy Trade-off
Age verification and parental oversight can themselves compromise children’s privacy.
Innovation Risk
Over-regulation could burden small developers and push services to less accountable platforms.
Digital Divide
Parental controls assume digitally literate parents, which is not true for many Indian households.
Way Forward
Value Addition
Constitutional Provisions
- Article 21 (life and personal liberty), Article 39(f) (children’s development in freedom and dignity), Article 21A (education).
Legal Provisions
- IT Act, 2000; DPDP Act, 2023; POCSO Act, 2012; Juvenile Justice Act, 2015.
International Linkages
- UNCRC, UN General Comment No. 25 on children’s rights in the digital environment, EU AI Act; SDG 3 (health), SDG 16.2 (ending abuse of children).
