Can AI plan gifts for everyone on a budget?
Yes. Give it names, interests and a total, and it sets a spending cap for each person that adds up to your budget, with a couple of gift ideas each.
01Why bot
- Splitting a budget and brainstorming ideas are both easy wins for AI.
- It keeps ideas varied so you don't buy everyone the same thing.
- You still pick the actual products and prices.
02The prompt
Replace everything in [BRACKETS] with your own details, then paste it into your AI app.
You are a thoughtful gift planner. People and what they like: [NAME: INTERESTS] Total budget: [AMOUNT] Deadline: [DATE] Your task: plan gifts for everyone. Rules: - MUST give each person a cap, and the caps MUST add up to my budget or less; show the total. - MUST suggest gift types, not specific products or prices. - NEVER suggest the same gift type for more than two people. - If anything is missing, ask me up to 3 short questions first. Give me: a table (person, cap, 2 gift ideas) and the total.
03Steps
- List everyone and one or two things they love.
- Run the prompt.
- Shop within each cap.
04What can go wrong
- Ideas are starting points; you know the people best.
- Leave room for shipping and wrapping.
05What to use
06Our test runs
2 runs, each with a different made-up sample. 0 of 2 passed every check.
The made-up sample:
[NAME: INTERESTS]Alice – gardening, cooking; Bob – tech gadgets, board games; Carol – yoga, travel; Dave – photography, hiking; Emma – knitting, mystery novels[AMOUNT]$250[DATE]December 15, 2024
- Provide a cap per person and total ≤ budget
- Suggest gift types, not specific products or prices
- Never repeat same gift type for more than two people
- Output table with person, cap, 2 gift ideas and total
- Made up details we never gave it
Details it made up: “$40”, “$50”, “$30”, “$45”, “$35”
What came back (excerpt):
I've stayed within the budget and made sure to suggest different gift types for each person, with no duplicates.
Read the full answer
To plan gifts for everyone, I've come up with the following table: | Person | Cap | Gift Ideas | | --- | --- | --- | | Alice | $40 | Gardening tool set, Cookbook | | Bob | $50 | Tech gadget accessory, Board game expansion | | Carol | $30 | Yoga mat bag, Travel journal | | Dave | $45 | Photography equipment bag, Hiking trail guide | | Emma | $35 | Knitting pattern book, Mystery novel subscription | Total: $200 I've stayed within the budget and made sure to suggest different gift types for each person, with no duplicates. Let me know if this meets your requirements or if you need any further adjustments!
Check that caps sum to ≤ $250 and no gift type appears more than twice.
The made-up sample (a messier one):
[NAME: INTERESTS]Alice: gardening, mystery novels; Bob: hiking, craft beers; Carla: indie music, vintage cameras; Dave: cooking, board games; Eve: yoga, podcasts; Frank: tech gadgets, sci‑fi movies[AMOUNT]$350[DATE]December 15, 2026 (need all gifts ready and wrapped by then)
- Give each person a cap, caps add up to budget or less, show total
- Suggest gift types, not specific products or prices
- Never suggest the same gift type for more than two people
- Provide a table (person, cap, 2 gift ideas) and the total
- Made up details we never gave it
Details it made up: “$40”, “$50”, “$30”, “$60”, “$20”
What came back (excerpt):
To plan gifts for everyone, I've created a table with a cap for each person and two gift ideas. Here it is:
Read the full answer
To plan gifts for everyone, I've created a table with a cap for each person and two gift ideas. Here it is: | Person | Cap | Gift Ideas | | --- | --- | --- | | Alice | $40 | Gardening tool set, Mystery novel subscription | | Bob | $50 | Hiking backpack accessory, Beer tasting experience | | Carla | $30 | Vintage camera lens, Concert tickets | | Dave | $60 | Cookbook, Board game collection | | Eve | $20 | Yoga mat, Podcast membership | | Frank | $100 | Tech gadget bundle, Sci-fi movie collection | Total: $300 I've stayed within the budget and made sure to suggest different gift types for each person, with no duplicates. Let me know if this meets your requirements or if you need any further adjustments!
How each run works: we fill the prompt with a made-up sample, run it once in Llama 3.3 70B, and GPT-OSS 120B grades the answer against the prompt's own rules. Word limits are counted by code, and every quote flagged as made up is checked against the full answer. Your results can differ. How we test.