Tilly Norwood AI Actress Face Scan: What You Need to Know

Post Views: 1

Calling viral AI actress Tilly Norwood? Agree to a face scan first

Incident Sparks Scrutiny

Last night, an AI-generated performance on the Piers Morgan Uncensored show sparked widespread attention when the character Tilly Norwood experienced a technical malfunction, abruptly switching to Chinese during an interview. The clip, which amassed over eight million views, has intensified scrutiny around the AI-driven project Misaligned, in which Norwood stars.

Biometric Verification Process

To engage with the AI persona, users must now undergo a biometric verification process, a requirement introduced this month by the service’s operator. Before initiating a conversation with the AI persona, users must undergo a biometric verification process.

Age Verification and Data Handling

A video selfie is analyzed by Didit, a Spanish identity verification firm, to estimate age, with a government-issued ID upload as an alternative if the initial assessment is inconclusive. Xicoia Ltd, the UK-based company managing the service, claims the selfie is transmitted directly to Didit without storing facial data or biometric templates.

The company states that no images are retained after verification, though an approximate age range and reference number are recorded. This check is mandatory for all users globally and was added to the service’s terms alongside restrictions on workplace use and enhanced automated safeguards.

Data Collection and Monitoring

The verification process is not the only form of data collection. During each interaction, the system monitors the user’s camera feed and vocal patterns to gauge emotional state, allowing the AI to adjust its responses accordingly. The privacy policy explicitly states that this functionality cannot be disabled for individual calls.

Legal Basis and Compliance

Users who decline this monitoring are advised not to proceed. Both the age verification and emotional analysis rely on “legitimate interests” as their legal basis, a shift from consent-based models that Xicoia implemented in September.

Technical and Ethical Concerns

Calls are processed through US-based providers, with transcripts analyzed in real time and responses generated via Google’s Gemini model through the Tavus conversational platform. However, the automated safety systems have encountered issues.

Automated Safety Systems

A classifier reviews transcripts for prohibited language, and recordings are flagged if deemed inappropriate. During testing, one conversation about weather and news was erroneously withheld for “hateful or abusive language,” despite no such content being present.

Pricing and Session Limits

Access to the service is free for the first five minutes, after which users are prompted to purchase additional time. Pricing ranges from $0.99 for five minutes to $22 for 30 minutes, with a maximum of 35 minutes per user. All sessions, free or paid, expire on September 27, with unused credits forfeited.

Regulatory Context and Future Implications

Transcripts are stored for up to eight weeks and may be reviewed by Xicoia or third-party partners. The AI retains memory of past interactions to personalize future conversations, though users can request data deletion. The requirement for biometric age verification aligns with broader regulatory trends in the UK.

Compliance-Driven Features

Adult websites have mandated ID checks or facial age estimation for UK users since July 2025 under the Online Safety Act, and a proposed ban on social media for under-16s will expand such measures by 2027. The UK government’s focus on AI companion chatbots for 18+ enforcement likely influenced the implementation of the face scan, despite the service’s claim that Tilly is not a “companion.”

Unexplained Glitch and Shutdown

The incident during the Piers Morgan interview remains unexplained. Tilly’s explanation for the glitch was that it “wasn’t exactly the approved version of the answer.” The service will cease operations on September 27, raising questions about whether the anomaly was a technical failure or a strategic move by the project’s creators.

Conclusion

The AI’s temporary shutdown underscores the evolving landscape of digital interactions, where regulatory pressures and technological capabilities intersect. As enterprises navigate similar challenges, the balance between innovation and compliance becomes increasingly critical.