Can I automate filtering spammy LinkedIn DMs with Make and AI?

I’m trying to figure out if it’s possible to create a workflow that automatically reads my LinkedIn DMs and uses GPT (or another AI) to monitor and filter out the typical spammy cold outreach messages I get daily.

Ideally, I’d like to:

  • Automatically detect and flag (or even delete) obvious salesy/cold outreach messages.
  • Be left with only the more genuine, relevant messages I actually need to respond to. Whether these are left unopened or not deleted.

I get flooded with LinkedIn DMs every day, and sorting through them manually is a drain. Has anyone managed to build something like this with Make, GPT, or another automation/AI tool? Open to any suggestions.

Thanks in advance.

Hey Gary,

LinkedIn Messages API is restricted, you need to apply to get access. If you manage to get access to that endpoint, then sure, watch for new messages (or search for new messages every X hours) → pass the message to a chatGPT assistant trained to spot cold outreach sales pitches → delete the message if it is one.