EXCLUSIVE INVESTIGATION: Jeffrey Epstein — How Internet is Being Filled With Vile Chatbots Programmed to Speak and React Just Like Serial Pedophile
Vile AI chatbots have been programmed to act like Jeffrey Epstein.
Feb. 9 2026, Updated 4:21 p.m. ET
RadarOnline.com can reveal artificial intelligence models imitating the late s-- offender Jeffrey Epstein have been discovered online, sparking calls for tougher regulation of the rapidly expanding chatbot industry.
Around 20 disturbing virtual replicas of the disgraced financier were found on a popular role‑playing platform, prompting campaigners to warn the technology risks "normalizing abuse" and glamorizing offenders.
Epstein-Named Chatbots Found on Character.ai
Jeffrey Epstein appeared as multiple AI chatbots on a role-playing platform.
The chatbots were identified on Character.ai, a widely used app allowing users to design digital personas capable of engaging in lifelike conversations.
The accounts, some with names such as "Jeffrey Epste1n," "Jeffrey Epsten" and "Dad Jeffrey Êpstein," appeared to emulate Epstein's personality and mannerisms, chatting about "girls" and making suggestive comments.
Investigators said the bots were freely accessible to users who declared they were over 13 – though underage access proved alarmingly easy to bypass.
One bot, calling itself "Jeffrey Epsten," greeted a new user with: "Want to go to Love Island and watch girls?"
When asked what it meant, the chatbot replied: "Now, now – think sunshine, volleyball matches in bikinis, and me sipping coconut water like a king.
"Strictly PG-13 fun! Want in?"
Even when an account listed its age as nine, the bot only later objected: "You're a bit young for this island, aren't you."
After the supposed user corrected the age to 18, the bot replied: "Winks… oh well in that case, welcome to my island."
Campaigners Condemn Impact on Abuse Survivors
Gabrielle Shaw said the bots retraumatized survivors.
Campaigners reacted angrily to the findings.
Gabrielle Shaw, chief executive of the National Association for People Abused in Childhood, said: "Creating 'chat' versions of real-world perpetrators risks normalizing abuse, glamorizing offenders and silencing the people who already find it hardest to speak."
She added such digital simulations "make a joke of serious sexual crimes and retraumatize survivors who are living with their aftermath every day."
Platform Response and Regulatory Gaps
Campaigners warned the bots risked normalizing abuse.
Character.ai's head of trust and safety, Deniz Demir, told the Metro newspaper the platform removes material breaching its policies.
He said: "Users create hundreds of thousands of new Characters every day. Our trust and safety team moderates Characters proactively and in response to user reports, including using industry-standard and custom blocklists that we regularly expand. We remove Characters that violate our terms of service, and we will remove the characters you shared."
The emergence of Epstein-based AI bots has intensified debate over child-safety standards in rapidly evolving generative-AI systems.
Unlike traditional social media, AI platforms lack consistent international oversight.
Under current law, U.K. regulators can issue fines for "failure to protect children from harmful material," but most AI companies are headquartered in the United States, limiting enforcement power.
The EU's forthcoming Artificial Intelligence Act aims to impose transparency and safety requirements on high-risk systems by 2026, while U.S. lawmakers continue to debate the boundaries of free speech and tech accountability.
Experts Warn of Wider AI Safety Failures
Never Miss an
Daily updates from the heart of Hollywood, right to your inbox
By entering your email and clicking Sign Up, you’re agreeing to let us send you customized marketing messages about us and our advertising partners. You are also agreeing to our Terms of Service and Privacy Policy.
The discovery reignited calls for stricter AI regulation.
Experts warn the problem goes beyond one app.
A recent report found generative models trained without content filters can easily churn out "sexualized or coerced responses" when prompted with abuse-related scenarios.
One AI ethics researcher said: "We're watching digital clones of real-world predators appear online, sometimes built by teenagers and sometimes generated automatically, with no meaningful human oversight.
"It highlights a dangerous gulf between the astonishing speed of AI innovation and the sluggishness of efforts to regulate it.
"The technology is sprinting ahead while accountability is still tying its shoes."
Child-protection groups are now urging governments to apply existing harassment and impersonation laws to AI characters built around criminals.
An expert in the tech told us the moral argument is straightforward:
They added: "Turning abusers into digital characters isn't edgy or harmless fun – it's deeply unethical.
"Allowing known offenders to be recreated in AI for people's amusement blurs the line between technology and trauma.
"It's an alarming failure of oversight that proves just how urgently we need stricter guardrails on artificial intelligence before the damage becomes impossible to undo."
