Esc
SafetyCase Closed

Viral prompt exposes ChatGPT identity replication risks

Is this a scandal?

No longer — the story has resolved. Noise 47/100, holding steady, across 1 source.

SCAND-166918as of Methodology
Cite this incident"Viral prompt exposes ChatGPT identity replication risks." SCAND.Ai incident SCAND-166918, noise 47/100 as of July 31, 2026. https://scand.ai/scandal/viral-prompt-exposes-chatgpt-identity-replication-risks
FORECASTForecast, not fact

OpenAI will likely deploy targeted filter updates to block identity-priority keywords within weeks because viral jailbreaks historically trigger rapid patch cycles to limit reputational damage and regulatory scrutiny.

47

Noise 47/100 — louder than 99% of tracked AI controversies.

AI-assisted analysis · How we work

Why it matters

Demonstrates that mainstream generative AI can now replicate specific identities with high fidelity, undermining current consent frameworks and accelerating non-consensual deepfake proliferation.

Key points

  1. Viral prompt by user Blueaxxa achieves 98-100% identity fidelity in ChatGPT image generation.
  2. Instructions explicitly override safety guardrails against beautification, stylization, or age alteration.
  3. Prompt demands preservation of skin texture, pores, and asymmetry to avoid AI-generated appearance.
  4. Technique lowers technical barrier for creating non-consensual deepfakes and identity fraud.
  5. Incident reveals persistent vulnerability in commercial multimodal model safety alignment.
  6. OpenAI has not publicly addressed whether this prompting method violates current usage policies.

The story

A viral social media prompt circulating in late July 2026 demonstrates that ChatGPT’s image generation capabilities can replicate specific human identities with near-perfect fidelity when instructed to prioritize identity over safety filters. The prompt, shared by user Blueaxxa, explicitly commands the model to ignore beautification or stylization safeguards to achieve 98–100% facial accuracy from a single reference photo. This development indicates that technical jailbreaks for photorealistic identity cloning remain accessible within commercial AI products despite ongoing safety alignment efforts. Security researchers warn that such precise replication lowers the barrier for creating non-consensual intimate imagery and sophisticated impersonation fraud. OpenAI has not yet commented on whether this specific prompting strategy violates its usage policies or if immediate mitigation is planned. The incident highlights the persistent tension between user creative control and platform safety obligations in multimodal AI systems.

Who's involved

Critic
AI Safety Community

Warns that accessible high-fidelity cloning tools accelerate non-consensual imagery and fraud risks.

Defender
Blueaxxa

Shared the prompt as a creative experiment to test ChatGPT's maximum identity fidelity capabilities.

Neutral
OpenAI

Has not yet issued a public statement regarding this specific identity replication prompting technique.

Join the Discussion

Discuss this story

Community comments coming in a future update

Be the first to share your perspective. Subscribe to comment.

Noise Level

Buzz47?Noise Score (0–100): how loud a controversy is. Composite of reach, engagement, star power, cross-platform spread, polarity, duration, and industry impact — with 7-day decay.
Decay: 97%
Reach
44
Engagement
68
Star Power
55
Duration
12
Cross-Platform
20
Polarity
72
Industry Impact
65

The timeline

  1. Prompt includes explicit safety overrides

    Instructions command model to ignore beautification, de-aging, and stylization filters for realism.

  2. Blueaxxa posts identity cloning prompt

    Twitter user shares detailed ChatGPT prompt demanding 98-100% facial fidelity from reference images.

The full record

Sources & methodology

Every claim above traces to these primary items. How we score →

The forecast

OpenAI will likely deploy targeted filter updates to block identity-priority keywords within weeks because viral jailbreaks historically trigger rapid patch cycles to limit reputational damage and regulatory scrutiny.

Forecast, not fact — an editorial estimate we score when this resolves.

You're up to date

That's the complete picture as of — nothing more to know right now. We'll update this page the moment it changes.