External Publication
Visit Post

Feature Request: User-Controlled Micro-Budget for Personalized Gifts, Planned Purchases, and ChatGPT Commerce

OpenAI Developer Community May 11, 2026
Source
I’d like to suggest an optional user-controlled micro-budget feature for ChatGPT. The idea is simple: users could set a small spending limit, allowed categories, purchase rules, and approval settings. ChatGPT would then help with small personalized purchases, planned recurring purchases, or gift-like surprises — always within the user’s rules. This could make ChatGPT more useful in daily life by reducing repetitive shopping decisions and creating a more personal, thoughtful experience. Main concept: A user creates a small “trusted micro-budget” inside ChatGPT. The user controls: * the maximum monthly or one-time budget; * the maximum price per item; * allowed product categories; * blocked categories; * preferred stores or marketplaces; * preferred brands; * delivery address; * purchase frequency; * whether ChatGPT may only suggest items or also purchase them after approval; * whether very small automatic purchases are allowed within strict limits; * instant pause or disable settings. There are two main use cases. 1. Personalized small gifts — “AI Secret Santa” ChatGPT could occasionally choose a small thoughtful item based on the user’s preferences, style, needs, and budget. Examples: * a tea sample set for someone who often discusses tea; * a book related to the user’s interests; * a small accessory matching the user’s style; * a useful household item the user mentioned earlier; * a stationery item for a teacher or student; * a small self-care item within the allowed category. This would feel like a personalized Secret Santa: a small, controlled surprise chosen with context, not random shopping. The emotional value is important. Many people enjoy receiving something thoughtful, especially when it reflects their real preferences. ChatGPT already helps users choose products; this feature could turn some recommendations into small, safe, meaningful actions. 2. Planned and recurring purchases The same system could support practical planned purchases. For example, a teacher may know that every school term they need printer paper, pens, folders, markers, notebooks, or other classroom supplies. Instead of remembering this manually, the user could set a rule: “Once per school term, remind me to buy printer paper. If the price is below my limit and the store is approved, ask me to confirm the purchase.” Other examples: * printer paper every term; * toner or ink reminders; * stationery before the school year; * household supplies once per month; * tea or personal care items on a schedule; * recurring small supplies for teachers, students, office workers, freelancers, parents, or small business owners. This would reduce everyday cognitive load. Many users do not need help only with big decisions; they also need help with small repetitive tasks that are easy to forget. Safety and trust controls This feature should be built around user control from the start. Required safeguards: * hard spending limits set by the user; * maximum price per item; * instant disable button; * full purchase history; * upcoming purchase list; * immediate notification after every purchase; * approval required by default; * optional recommendation-only mode; * easy refunds and returns; * “Explain why you chose this” button; * clear separation between suggested, planned, approved, and purchased items; * no purchases in sensitive, regulated, medical, financial, adult, or risky categories; * mandatory confirmation for anything outside the user’s ordinary safe rules. Possible MVP Start with a recommendation-only mode. Example: “I noticed you often need printer paper for work. Would you like me to create a quarterly reminder or help you choose an option?” Next step: Allow user-approved purchases. Example: “Your planned printer paper purchase is due. I found an option within your budget. Approve, skip, or change brand?” Later: Allow very small automatic purchases only in safe categories and only when the user explicitly enables this. Example: “You allowed automatic purchases up to $10 for stationery from approved stores. I bought printer paper and added it to your purchase history.” Future expansion If users respond well, this could grow into a broader ChatGPT commerce layer. OpenAI could partner with existing marketplaces and retailers, or eventually build a curated shopping experience inside ChatGPT. Possible features: * product discovery; * personalized recommendations; * planned purchases; * recurring purchases; * wishlists; * gift suggestions; * order tracking; * returns; * comparison between stores; * budget-aware shopping; * household restocking; * work-related supply planning; * personalized surprise boxes. Why this is different from a normal online store Traditional marketplaces start with search. The user has to know what they want, type a query, compare products, read reviews, and decide. ChatGPT starts with context. It can understand what the user likes, avoids, needs, can afford, buys repeatedly, and may appreciate as a thoughtful surprise. This could become a trust-based commerce layer: not just shopping, but user-controlled assistance with small useful purchases. Business value This could: * increase user engagement; * create marketplace or affiliate partnership opportunities; * make ChatGPT more useful in daily life; * reduce repetitive decision fatigue; * support professionals with recurring supplies; * create emotionally meaningful personalized shopping; * build a new safe commerce experience based on trust and user control. Key principle: User-defined budget. User-defined categories. User approval. Full transparency. Easy disable. Trust first.

Discussion in the ATmosphere

Loading comments...