On 20 January 2026, OpenAI announced it is rolling out age prediction technology across ChatGPT globally. The system uses behavioural analysis to identify users under 18 and automatically restricts their access to sensitive content, including graphic violence, sexual material, and content promoting self-harm.
The announcement, detailed in OpenAI's official blog post, comes as AI platforms face increasing scrutiny over their interactions with young users. As Engadget reports, the move prepares the platform for an upcoming "adult mode" that will allow verified adults to access content currently restricted for all users.
Who Is Affected
The change affects all 800 million weekly active ChatGPT users worldwide. Specifically:
- Users identified as under 18 will see a restricted version of ChatGPT that limits sensitive content
- Adults incorrectly flagged can verify their age through identity verification service Persona
- Business users on ChatGPT Team, Enterprise, and API plans are not affected by these consumer-facing controls
- Free and Plus subscribers will see the changes first, with the Go tier following
For UK businesses, this represents a significant shift in how the most widely-used AI platform manages content. If your employees or customers use ChatGPT, they may notice changes in the responses they receive.
How the Age Prediction System Works
According to OpenAI, the age prediction model analyses "a combination of behavioural and account-level signals" to estimate user age:
- How long an account has existed
- Typical times of day when someone is active
- Usage patterns over time
- The user's stated age during signup
The system does not use facial recognition or require identity documents for the initial prediction. However, users who believe they have been incorrectly classified as minors can submit identification through Persona, which OpenAI states will delete the verification data within seven days.
What Content Is Restricted for Under-18s
Users identified as minors will encounter restrictions on:
- Graphic violence and detailed descriptions of harm
- Sexual or romantic roleplay content
- Content relating to self-harm or eating disorders
- "Risky viral challenges" and dangerous activities
- Extreme beauty standards or unhealthy dieting advice
These restrictions apply automatically and cannot be disabled by users identified as under 18.
Why This Matters for AI Visibility
This development signals something important about the trajectory of AI platforms: they are maturing rapidly and implementing sophisticated content governance systems.
For businesses focused on AI visibility, three implications stand out:
1. AI Platforms Are Becoming More Regulated
OpenAI's move reflects broader regulatory pressure on AI companies worldwide. The company is simultaneously expanding ChatGPT Edu access to entire countries, suggesting a dual strategy of growth and governance. The UK government is considering tighter rules for AI chatbots, with Technology Secretary Liz Kendall expressing concern about young people forming "unhealthy relationships" with AI tools. California's companion chatbot regulations took effect on 1 January 2026.
Businesses that want AI platforms to accurately represent them need to understand that these platforms are implementing increasingly complex content policies.
2. Behavioural Analysis Is Now Standard
The fact that ChatGPT now analyses user behaviour patterns to make decisions about content delivery demonstrates how sophisticated these systems have become. This same technology that identifies user age could theoretically influence how AI systems respond to different types of queries about businesses.
3. Content Differentiation Is Coming
OpenAI is preparing for differentiated content tiers. The upcoming "adult mode" suggests that AI platforms may increasingly serve different content to different user segments, and with ChatGPT now preparing to introduce advertising, monetisation adds another layer of complexity. For businesses, this raises questions about how their products and services might be represented to different audiences.
"AI safety is genuinely hard."
- Sam Altman, CEO of OpenAI, responding to criticism about ChatGPT safety
This acknowledgement from Altman reflects the real tension AI companies face: building systems that are both useful and safe. In my work helping businesses establish their AI visibility, I have seen how this tension plays out practically. Businesses want AI to recommend them confidently, but AI systems are becoming more cautious about what they say and to whom. Understanding this balance is increasingly important for anyone who wants their business to be accurately represented in AI responses.
The Broader Context
OpenAI's announcement comes amid heightened scrutiny of AI chatbots and their impact on young users. Several high-profile cases have linked AI interactions to harm, and regulators globally are responding.
In the United States, the Federal Trade Commission opened an inquiry into AI companion chatbots in September 2025, specifically citing concerns about their influence on "vulnerable users, particularly children and teens."
The UK government is weighing tighter regulations, with concerns that AI chatbots may "expose children to harmful content or encourage self-harm." This follows a case where a teenager's death was linked by his family to interactions with an AI chatbot.
"While adults should have the freedom to use AI tools more openly, children should be more carefully protected."
- Sam Altman, during Senate testimony on AI capabilities
When I first read this statement from Altman, I was struck by how it acknowledged the fundamental challenge facing AI platforms: they cannot simply treat all users identically. This has broader implications than just child safety. It suggests that AI platforms are moving toward more nuanced, context-aware interactions. For businesses thinking about AI visibility, this means the AI systems they want to be represented in are becoming more sophisticated about who they are talking to and what they should say.
Practical Implications for Businesses
While this update primarily affects consumer safety, businesses should consider several factors:
Employee AI Usage
If your team uses ChatGPT for work, be aware that usage patterns are now being analysed. This does not affect the privacy of conversation content, but it does mean that the platform is building user profiles based on behaviour.
Customer-Facing AI
If you direct customers to use ChatGPT or similar tools, understand that their experience may vary based on how the platform categorises them. A young customer researching your products may receive different information than an older one.
AI Visibility Strategy
As AI platforms implement more sophisticated content controls, the importance of clear, accurate information about your business increases. AI systems that are uncertain about what to say may default to saying less. Businesses with well-structured AI visibility implementations give these platforms confidence to speak accurately about them, but measuring that visibility accurately is another challenge, since many AI visibility checkers are not giving you real results.
What to Watch
Several developments in the coming months will clarify the broader implications:
- Adult mode rollout: OpenAI has not announced timing, but the age prediction system is clearly preparation for tiered content access
- UK regulatory response: The government's deliberations on AI chatbot regulation could influence how platforms operate in the UK market
- Other platform responses: Google's Gemini, Anthropic's Claude, and Microsoft's Copilot may implement similar systems
- Business API implications: Currently, business tiers are unaffected, but future updates may introduce content controls for enterprise users
Frequently Asked Questions
When does the ChatGPT age prediction system take effect?
OpenAI announced the global rollout on 20 January 2026. The system is being deployed progressively, so users may see it enabled at different times depending on their region and account type.
Does this affect ChatGPT business accounts?
Currently, ChatGPT Team, Enterprise, and API users are not subject to the consumer age prediction controls. However, OpenAI may introduce separate guidelines for business use in future updates.
How accurate is the age prediction system?
OpenAI has not published accuracy figures. Adults who are incorrectly identified as minors can verify their age through identity verification service Persona. The company states that verification data is deleted within seven days.
Does OpenAI store personal age data?
The age prediction uses behavioural signals and account information rather than requiring identity documents. For users who choose to verify their age, OpenAI's partner Persona handles the verification and deletes the data within seven days.
Will other AI platforms implement similar systems?
Given regulatory pressure and OpenAI's market-leading position, it is likely that Google Gemini, Anthropic's Claude, and Microsoft Copilot will implement comparable age verification measures. No announcements have been made as of this writing.
How does this affect AI visibility for businesses?
The immediate impact on business AI visibility is minimal. However, the development signals that AI platforms are implementing sophisticated content governance, which may influence how they respond to queries about businesses in future updates.
Understand Your AI Visibility
As AI platforms become more sophisticated, understanding how they represent your business becomes increasingly important. Our AI Visibility Checker shows you exactly what ChatGPT and other AI systems currently say about your business.
Check Your AI Visibility
