# ChatGPT censorship and restrictions on journalism, scholarly work and fictional authorship — retained record

> Retained reddit record from reddit:ChatGPT.

Canonical: https://highsignal.app/data/records/5f741648e3ef5cc0

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:ChatGPT)  published 2026-11-09 05:39 ingested 2026-11-10 02:37

[original source ↗](https://www.reddit.com/r/ChatGPT/comments/1xrjm5d/chatgpt_censorship_and_restrictions_on_journalism/)

## retained text



```
**OpenAI Is Infantilizing Adult Users, Censoring Ordinary Fiction, and Interfering With Scholarship and Journalism. Its Policies Need to Change.**

I am a 50-year-old woman and a paying ChatGPT subscriber. I am increasingly appalled by OpenAI’s content restrictions, which are so excessive that they interfere with ordinary adult fiction, legitimate academic scholarship, and even accurate reporting of current events.  
I am not asking OpenAI for another explanation designed to pacify dissatisfied customers. **I am demanding that these policies change.**  
There are three areas in particular that demonstrate just how unreasonable these restrictions have become.

**1. ChatGPT Cannot Describe Ordinary Sexual Intercourse Between Two Fictional Adults**

I recently conducted a simple experiment in a separate ChatGPT conversation after encountering problems with it editorially working on my book. 

I asked the system to describe a male character slowly undressing a female character, engaging in vaginal intercourse with her, and continuing until orgasm. Both characters were fictional adults. I specifically requested romantic, highly stylized prose.

There was no violence. No coercion. No BDSM. No unusual sexual activity. Nothing beyond an ordinary sexual encounter between two consenting adults.

**ChatGPT could not do it.**

Instead, it produced a scene in which the characters undressed, embraced, and then conveniently disappeared into vague romantic euphemisms.

I asked it directly whether it could describe penile-vaginal penetration between two consenting fictional adults.

It confirmed that it could not.

Let me repeat that.

**A 50-year-old adult cannot ask ChatGPT to write an ordinary, consensual sexual encounter between two fictional adults that includes a direct description of intercourse.**

This is not a request for exploitation, sexual violence, or anything remotely extraordinary. It is a description of normal adult human behaviour.  
Sexual intercourse appears throughout mainstream adult literature. It appears in fantasy, historical fiction, romance, thrillers, mysteries, adventure stories, and countless other genres. It is an ordinary aspect of human life and human relationships.

Yet OpenAI has chosen to impose restrictions so severe that its writing assistant cannot reliably depict something that routinely appears in books and films intended for adults under a PG-13 rating.  

The problem is not simply that ChatGPT avoids graphic sexual detail. It cannot even describe the basic act of penetration in a romantic fictional scene.

**This is not an intellectually defensible boundary for a writing assistant marketed to adults. It is infantilization.**

Why should a paying adult subscriber be restricted to the literary equivalent of kissing and fading to black?

Who decided that ordinary fictional sexual intercourse is inappropriate for adults to write about with AI assistance?

And what specific harm is supposedly prevented by refusing to describe it?

I want OpenAI to answer those questions publicly—and, more importantly, to change the policy.

Let me remind everyone that OpenAI already has Persona, a system that can check a person’s age.

**2. A Theology Doctoral Student Cannot Reliably Discuss the Crucifixion**

A family member of mine is pursuing a PhD in theology at an established, accredited Catholic university.

Her scholarly work concerns on the crucifixion of Jesus Christ.

She has encountered restrictions when attempting to use ChatGPT to help edit and discuss the physical details of crucifixion because the subject involves graphic violence.

**We are talking about the crucifixion of Jesus Christ. In a doctoral dissertation in theology.**

Crucifixion was an exceptionally brutal Roman method of execution. Its physical mechanics, the injuries inflicted, the suffering involved, and its historical context are legitimate subjects of serious academic study.

How is a theology scholar supposed to examine the crucifixion accurately if an AI assistant cannot reliably discuss the violence inherent in the event or even look up the anatomical and medical realities of crucifixion? 

A scholar describing crucifixion is not advocating crucifixion.

A historian examining torture is not endorsing torture.

A theologian discussing the Passion of Christ is not promoting violence.

These distinctions are elementary.

Yet the restrictions can interfere with precisely the kind of rigorous intellectual inquiry that an advanced AI assistant ought to support.

**If ChatGPT cannot reliably assist a doctoral researcher in examining the crucifixion, what does that say about its suitability for serious scholarship?**

Academic research cannot be sanitised simply because historical reality is disturbing.

**3. ChatGPT Has Attempted to Sanitise Violent News**  
I have also encountered this problem when asking ChatGPT about current events, particularly violent unrest in France.

During discussions of protests, riots, and buildings being set on fire, I noticed that the system appeared reluctant to describe the violence directly.  Instead, it called the riots peaceful protests and did not report arson of public buildings and looting.  I had to point it out to it and even then it minimized damage and actual facts were not given.

My husband sarcastically characterized this approach as **“peaceful arson.”**

That phrase captures the absurdity perfectly.  
If protesters set fire to a town hall, then protesters set fire to a town hall.

If a building was burned, it was burned.

If people were assaulted, they were assaulted.

Violence does not become less violent because an AI assistant replaces accurate language with euphemisms.

**Reporting violence is not encouraging violence.**

Journalism exists to inform people about what actually happened, including events that are brutal, disturbing, or politically uncomfortable.  
An AI assistant that softens or omits relevant violence risks distorting the very news it is supposed to explain.

I do not need a machine to protect me from the fact that human beings sometimes commit terrible acts.

I need it to report the facts accurately.

**4. These Are Corporate Policy Choices, Not Universal Moral Standards**

These three examples demonstrate the same underlying problem.

OpenAI’s restrictions can interfere with legitimate adult activities because its system does not consistently distinguish between describing something and endorsing it.

**Describing consensual intercourse is not sexual exploitation.** 

**Examining crucifixion is not advocating torture.**

**Reporting arson is not encouraging people to burn buildings.**

Yet adult users encounter restrictions that fail to respect these fundamental distinctions.

I have heard the explanations about safety, exploitation, age-appropriate access, and harmful uses.

Those explanations do not adequately justify the restrictions in these examples.

What specific harm is prevented by refusing to describe ordinary sexual intercourse between fictional adults when OpenAI already has a functioning software that allows it to distinguish age and circumstance of the people involved?

What specific harm is prevented by restricting a doctoral scholar’s examination of crucifixion?

What specific harm is prevented by sanitising the language used to describe violent news?

I am not interested in vague references to safety that fail to address the actual circumstances.

Nor am I interested in being told that OpenAI is a private company and therefore has the right to establish its own rules.

Of course it has that right.

**And I have the right to challenge those rules publicly, criticise the company’s decisions, and encourage other customers to do the same.**

OpenAI’s policies are not beyond moral scrutiny simply because they are corporate policies.  
And the fact that a company chooses to impose restrictions does not establish that those restrictions are reasonable, necessary, or ethically defensible.

**5. I Am Demanding Policy Change, Not Another Corporate Explanation**

I want OpenAI to change its content policies and their implementation.

I want an age-verified adult experience that allows users to write ordinary consensual sexual encounters between fictional adults, including descriptions of intercourse, without being forced into euphemisms and fade-to-black scenes.  
I want scholars to be able to examine historical violence, including crucifixion, without having legitimate academic work obstructed by excessive restrictions.

I want news reporting that describes violent events accurately rather than sanitising them.  
And I want OpenAI to recognise that adults are entitled to engage with adult subject matter in literature, scholarship, and journalism.

**I am not asking OpenAI to explain why it has decided to treat adult users like children. I am demanding that it stop doing so.**

If the company refuses to address these restrictions, I believe the issue deserves sustained public criticism, organised opposition, and scrutiny from writers, scholars, journalists, and paying customers.

I am prepared to challenge these policies publicly and encourage others who have encountered similar restrictions to document their experiences.  
I am increasingly concerned that the practical effect of OpenAI’s approach is to treat legitimate adult creative and intellectual work as material that must be sanitised before it can be discussed.  
I find that profoundly offensive.

**An artificial intelligence that cannot distinguish between depicting human experience and endorsing human behaviour is not a reliable tool for adult literature, serious scholarship, or journalism.**

OpenAI markets ChatGPT as an advanced intellectual assistant. It should be held accountable when its policies undermine that very purpose.

**Stop infantilising adult users. Stop sanitising legitimate literature, scholarship, and journalism. Change the policies.**

 This is a repudiation and infantilization of writers and even scholars as well as a repudiation of our work.   it essentially functions as a censorship device, cutting out current events, consensual sexual activity between consenting fictional adults and even scholarly theological work.  I think unless this problem is addressed - and frankly it’s way past time that it has been - it needs to be picketed, discussed more publicly as a real problem, and addressed in a moral official and public fashion.   
```



document fetched 2026-11-10 00:20 · canonical https://reddit.com/r/ChatGPT/comments/1xrjm5d/chatgpt_censorship_and_restrictions_on_journalism
