Unpopular opinion
retained text
Before hand I want to state that im absolutely pro AI and I use It extensively. In university, Coding, Hardware development and at work (for automations(trying)). I am using the top models of Anthropic and Open AI. It's probably my content bubble but wherever I look I see People claiming how great everything works when they are automating all their processes and one shotting some useless Ai slop. And I don't get it. Don't get me wrong it's crazy how far these models have come in coding and other task but F me are they bad in a huge amount of other things. I recently tried (3 times, also talking with strawberry browser directly on how to best do it) to automate an Emailing process with strawberry browser, writing emails and updating a google sheet using the google extensions. It's ok for in the setup and the first try but, it's extremely error prone and in the long run setting it up and monitoring it is way more exhausting and time consuming than just doing it the old fashioned way. Also for my coding and hardware development which includes a large repo and real life test. It helps me tremendously but still the lack of the most basic real world understanding of those models is sometimes mind blowing. All that combined with the general inflexibility of those models which are designed to just power throug issues as if they would not exist. And yet here we are with seemingly most people stating how PHD lvl intelligent those models are. And again I dont get it, powerful yes but intelligent, absolutely not. I always find my self imagining if I would work with a PHD student that acts and talks like a model, I would probably be in jail already. And last but not least regarding the current talk about a super intelligent AI killing us all - I think that could actually happen but not because its too intelligent but because its too stupid to understand when to stop. Am I the only one who does not get how these marketed automations are supposed to actually work without giving me a heart attack and more work.
document fetched 2026-10-25 00:19 · canonical https://reddit.com/r/AI_Agents/comments/1xe5ard/unpopular_opinion