Hi Everyone,
I wanted to share something that I’m hoping will help me when it comes to making better choices about what I buy. While this isn't about promoting European products in the first instance, it's designed to help me figure out which brands I currently use, but may need to avoid, so I can find better alternatives that are local, European, allied countries, or even values-led US brands if no other option exists.
Here's what I did:
-
Got the idea from Reddit: I saw a post on Reddit with a visual of all the brands owned by big multinational companies. Really helpful to see the visual but there were literally thousands, and I wanted a text list I could feed into an AI chat as a reference document.
-
Did Some Research: I looked up these multinational companies on Wikipedia which seemed to have a relatively updated list of subsidiary brands, focusing on the parent companies I wanted to steer clear of, like PepsiCo, Nestle etc.
-
Cleaned Up the List: I used AI tool to organise everything into a neat PDF.
-
Made It Easy to Use for me with AI: I added the PDF to a chat in Mistral so I can quickly check any products I see in my fridge or cupboard. The AI helps me spot any brands I should avoid. I can also just list the brands off using voice for ease/speed but accuracy suffers a bit. I may also be able to just take a picture but haven’t tried that yet.
The image is the output.
Anybody with access to a simple list and an LLM could do this. Been inspired by the excellent websites and resources others have made and posted here, which I will use to find alternatives - but I first needed to know where the problems were!
Im hoping this will make it easier to buy stuff that aligns with my values. Hope this makes sense and helps someone else too!
While I agree the environmental impact of LLM is concerning, I disagree that this is a poor use for it.
It's pretty much the best case scenario for using an LLM. There are so many companies within companies that if you're trying to avoid an umbrella brand an LLM can quickly parse through that data and provide a response. Whereas you may take 5 minutes trying to find an answer and if you're grocery shopping that could add up very quickly.
And this is trivial for an LLM to do and would use barely any compute. As always, companies that are using AI to generate images, replace staff and moderate their websites that are going to be creating 99% of the environment damage by AI. We shouldn't be going after individuals who are using it to make their day-to-day easier especially if they're using it to hurt monopolies.
Correct me if I'm wrong, but if I understand correctly from the steps provided, it was manual data extraction.
AI was used to "organize data into a neat PDF", and to later ask the AI whether a brand is okay to use or not, with a PDF that presumably contains "a list of brands to avoid" according to the prompt. This can literally be a Ctrl+f or something similar.
I'm not going after anyone. Yes, we should be going after corporations, and not only for their environmental footprint, this doesn't mean we can't criticize other people.
Plus, I really agree with the sentiment, as already stated. But, imo, this is not the right way to go.
May help if I clarify. The document is ingested into the chat, so acts as a local knowledge base. When I put some random brands into the llm it uses the info from the doc to respond.
I probably could manually search the document (ctrl-f), just find it easier to use the llm approach, especially if inputting in an unstructured/messy way.
Does the llm search itself for subsidiaries of the brands on the PDF on the net? Because it could then be a good use of AI to find the one you stumble upon in the millions of brands big players have participation in...
not deliberately, however when run it did automatically identify a brand (Hellman’s) owned by a multinational NOT in the list (Unilever). I hadn’t included Unilever in the source doc as it is British/Dutch. So it must have used its own knowledge base.
Perhaps it interpreted my preference as wishing to avoid multinationals? This is what it output:
“Based on the list you provided and the document, here are the brands you should avoid if you are following the list:
The other brands you mentioned are not listed in the document as brands to avoid. However, it's always good to double-check with the latest information or packaging details, as brand ownership can change.”
Can't you just search the pdf instead
Are you a bot? You literally replied to a comment mentioning that umbrella corps are hard to find and requires a few minutes of searching, whereas an LLM already knows it.
I mean that's true, but in this case, the OP downloaded a pdf with all the brands and umbrellas they wanted to avoid and gave it to the LLM.
I agree that an AI could be more intuitive and easy to use.
My point i guess is this is not something an AI just solves on its own and that you couldn't solve without an AI
yes you probably could just search ctrl+f. I'm quite lazy so imagined that if I could upload the document (46 pages PDF of companies and their brands), I could then just chat with the bot/or perhaps take a picture of a fridge/products and it could quickly give me the answer.