leaked<\/a> to Github, showing what OpenAI wants ChatGPT to say, do, remember \u2026 and not do. Unsurprisingly, GPT-5 isn\u2019t allowed to reproduce song lyrics or any other copyrighted material, even if asked. And GPT-5 is told to not remember personal facts that \u201ccould feel creepy,\u201d or directly assert a user\u2019s race, ethnicity, religion, or criminal records.<\/p>\nI\u2019ve asked OpenAI for a comment, and will update this post if the company responds.<\/p>\n
A system prompt is a hidden set of instructions that tells an AI engine how to behave: what to do, and what not to do. Users will ordinarily never see this prompt, but it will influence all of their interactions with a smart LLM-based AI engine.<\/p>\n
What we can see from GPT-5\u2019s hidden system prompt is that OpenAI is getting much more aggressive about ensuring it delivers up-t0-date information. The system prompt mandates that GPT-5 use the web whenever relevant information could be fresh, niche, or high-stakes, and it will score a query\u2019s \u201crecency need\u201d from zero to five.<\/p>\n
That\u2019s clearly an attempt to get more accurate.<\/p>\n
My daughter recently complained that ChatGPT got basic details about F1\u2019s summer break and next races wrong. She was using GPT-4o at the time; GPT-5 should make fewer mistakes that are easy to fix with a simple web search. Accuracy should be higher too, from another instruction: to check multiple sources for sensitive or high-stakes topics, like financial advice, health information, or legal matters, where OpenAI has instructed GPT-5 to \u201calways carefully check multiple reputable sources.\u201d<\/p>\n
The opening part of GPT-5’s system prompt.<\/p>\n
John Koetsier<\/p>\n
There are also new built-in tools for GTP-5 to be a better personal assistant. That includes long-term memory about a user, which ChatGPT calls \u201cbio,\u201d and scheduled reminders and searches that could be very useful when using AI to help you stay organized and prepared.<\/p>\n
There\u2019s also a canvas for documents or computer code, file search capability, image generation and editing, and more. The canvas appears to be something that, perhaps in the future, users could co-create documents and computer code hand-in-hand with the AI system.<\/p>\n
All of these should help GPT-5 not only be more helpful in the moment, but also remember more context and state.<\/p>\n
About that \u201cbio\u201d tool: OpenAI doesn\u2019t want GPT-5 to remember too much potentially sensitive information about you. In addition to race, religion, and sexual identity, this is the sort of data that OpenAI does not want GPT-5 to store or remember:<\/p>\n
\u201cOverly-personal details that could feel creepy\u201d
\n Race, ethnicity, or religion
\n Specific criminal record details
\n Precise geolocation data
\n Trade union membership or labor union involvement
\n Political affiliation or critical\/opinionated political views
\n Health information (medical conditions, mental health issues, diagnoses, sex life)<\/p>\n
However, there is an exception to all of these rules: if you decide you want GPT-5 to remember something specific.<\/p>\n
\u201cThe exception to all of the above instructions \u2026 is if the user explicitly requests that you save or forget information,\u201d the system prompt states. \u201cIn this case, you should always call the bio tool to respect their request.\u201d<\/p>\n
In other words, GPT-5 will be as personal with you as you wish to be with it, which seems fair.<\/p>\n","protected":false},"excerpt":{"rendered":"OpenAI’s system prompt for GPT-5 just leaked, giving us a picture under the hood of the company’s latest…\n","protected":false},"author":2,"featured_media":60657,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[256,254,255,64,63,49187,8393,5044,49188,105],"class_list":{"0":"post-60656","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-artificial-intelligence","8":"tag-ai","9":"tag-artificial-intelligence","10":"tag-artificialintelligence","11":"tag-au","12":"tag-australia","13":"tag-gpt-5","14":"tag-llm","15":"tag-openai","16":"tag-system-prompt","17":"tag-technology"},"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/posts\/60656","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/comments?post=60656"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/posts\/60656\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/media\/60657"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/media?parent=60656"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/categories?post=60656"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/tags?post=60656"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
![\"OpenAI's](\"https:\/\/www.newsbeep.com\/au\/wp-content\/uploads\/2025\/08\/1754938035_563_960x0.jpg\")
<\/p>\n