This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
ChatGPT experienced service disruptions where specific names triggered hard-coded filters, causing the system to terminate conversations and return error messages. While OpenAI attributed some instances to 'mistaken' tool flagging, the behavior raised concerns about the use of blunt, rule-based interventions to manage defamation risks and the potential for these filters to be exploited or to negatively impact user experience.
ChatGPT has reportedly been experiencing errors and service disruptions caused by hard-coded filters designed to prevent it from producing potentially harmful or defamatory content about certain individuals by blocking prompts containing specific names, likely related to post-training interventions. The reported names are Brian Hood, Jonathan Turley, Jonathan Zittrain, David Faber, David Mayer, and Guido Scorza.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
AI systems that fail to perform reliably or effectively under varying conditions, exposing them to errors and failures that can have significant consequences, especially in critical applications or areas that require moral reasoning.
Also classified as
The underlying cause for the filters was the AI system's tendency to generate false, defamatory hallucinations about real individuals.
OpenAI implemented hidden, blunt blocklists without transparency, leaving users and affected individuals confused about why certain names broke the system.
The AI system failed to perform reliably, crashing and terminating sessions when encountering specific names, and proved vulnerable to adversarial prompt injections exploiting these hard-coded rules.
Human
Due to a decision or action made by humans
Intentional
Due to an expected outcome from pursuing a goal
Post-deployment
Occurring after the AI model has been trained and deployed
The service disruptions were caused by a deliberate human engineering decision to implement hard-coded, rule-based filters to block specific names post-deployment.
Added categories peak at 2: Minor (Epistemic Harm, Direct)
OpenAI implemented hard-coded filters in ChatGPT to block specific names to prevent defamation lawsuits, leading to service disruptions and vulnerability to prompt injections. This incident represents a commercial software robustness and safety issue with negligible national security implications.
The reports describe ChatGPT as a 'generative AI chatbot' and 'AI language model' that responds to user prompts and queries, acting as an assistant to users.
Limited Risk: ChatGPT is a chatbot, which falls under Risk Level 3 (Limited Risk) requiring specific transparency obligations so users know they are interacting with an AI.
ChatGPT is primarily deployed as a general-purpose conversational chatbot designed to answer user questions and interact in natural language.
The reports from reputable sources (Ars Technica, WSJ, NYT) provide consistent, detailed accounts of the name-blocking behavior, the specific names involved, the reasons behind the blocks (defamation concerns), and OpenAI's response confirming the glitch.
None
Technology
Blunt Rule-Based Filters
Hard-coded 'if-then' rules used to override LLM outputs abruptly.
Black Box Neural Networks
Lack of understanding of LLM inner workings forces reliance on blunt rules.
LLM Hallucinations
AI generating false, defamatory claims about real people.
Data Inputs
Insufficient Training Data
Lack of accurate info about individuals leads to AI confabulations.
Name Ambiguity in Data
Failure to distinguish between different people sharing the same name.
Human Factors
Adversarial Prompt Injection
Users exploiting the filter using hidden text or visual prompt injections.
Process and Methods
Inadequate Testing of Tools
Automated tools mistakenly flagged benign names as privacy violations.
Blunt Data Redaction Methods
Using complete name bans instead of precise information filtering.
Management
Lack of System Transparency
Failure to communicate block lists and moderation reasons to users.
Reactive Safety Measures
Implementing hasty blocks in response to legal threats.
Regulatory Environment
Strict Defamation Liabilities
Lawsuits forced rapid, poorly-tested filtering of specific names.
Right to be Forgotten Laws
EU privacy regulations pressure platforms to remove personal data quickly.
AI system involved
Identified by automated extraction; can differ from the alleged developers above.
