5 thinking will hide its chain of thought completely when discussing sensitive topics or areas you told it previously not to reference
Understanding the Behavior of AI Language Models When Addressing Sensitive Topics In the landscape of…
You ever pull this crap again, I’m cancelling my account and just going to a free model
Evaluate Your AI Service Subscriptions Carefully: Ensuring Value and Transparency In the rapidly evolving landscape…
The GPT-4o to GPT-5 Rerouting is the Perfect Mirror of Our Toxic Fear-Based Society
The Transition from GPT-4o to GPT-5: A Reflection of Society’s Fear-Driven Culture In recent developments…
Did ChatGPT 5 become incredibly condescending in the past 24 hours or is it just me?
Recent Changes in ChatGPT Behavior: A Closer Look at User Experience In recent days, many…
Whatever is going on with 4o, it’s also affecting GPT-4.5
Examining Recent Developments in AI Model Behavior: The Impact on GPT-4.5 In recent days, users…
How ChatGPT 5 can feel more like 4o and both versions can be more stable.
Enhancing Stability and Personalization in ChatGPT: Strategies for a Consistent AI Experience As artificial intelligence…
Automatic routing. When a conversation is flagged internally as “complex” (lots of tokens, layered role-play, sensitive themes, or high safety risk), the servers don’t keep it on 4o. They route it to GPT-5, because 5 is newer, more aligned, and has more built-in safety features.
Understanding Automatic Conversation Routing to GPT-5: Implications and Concerns In the evolving landscape of AI-driven…
System prompt confirms: OpenAI Knows 4o reroutes “Sensitive” prompts to 5 (Screenshot from X)
Investigation Reveals Hidden Model Routing in OpenAI's 4o System: Is Sensitive Content Being Automatically Escorted…
ChatGPT switching thinking models while I have set it “instant”
Understanding ChatGPT’s Dynamic Model Switching: Insights and Troubleshooting In recent discussions within the AI community,…