korrents
Brian Lovin

What Brian Lovin thinks about LLMs

@brian-lovin · 20 positions · 0 changes of mind

Product designer at Notion working on AI products, and maker of the read-it-later app Shiori.

Everything they publish, on ppll ↗

Brian Lovin did not write this page.

We collected these quotes from things they published elsewhere, and every quote links to where it was said. They have no account here and have not endorsed this site. Quotes are word for word; the short line under each one is our own restatement, not their wording. Their own site. Is this you? Claim it or ask us to remove it. Or tell us what is wrong here.

10 dated positions, 2026, in their own words. Our reading of what Brian Lovin has said — not written or endorsed by them.

  1. Our readingDesigners should be designing right at the boundary of the current model, hoping the next one solves it; understanding what is possible and pushing the edges even if you get stuck is probably a good place to be right now.

    designvibe coding

    Their wordsI think where designers need to be designing right now is like right at the the boundary of the next the current model, like hoping that the next one solves it, right? Like we're all shipping shitty vibe-coded slop just hoping like, "Ah, the next model's definitely going to clean this up, right?" And I think that's probably a good a good place to be right now as a designer is really understanding what's possible, trying to push the edges even if it sucks and you get stuck

    ↗Brian Lovin - How to level up with AI as a designeryoutube.com 2nd of 20 in this recording

  3. Our readingPeople trying to use AI to automate knowledge work or simplify their job are very early adopters, and their tolerance for sub-par experiences is quite high.

    Their wordsif you're really trying to use AI right now to automate knowledge work or simplify your job or like remove a process at your company, you're a very early adopter and the tolerance for those people for like kind of sub-par experiences is actually quite high.

    ↗Brian Lovin - How to level up with AI as a designeryoutube.com 3rd of 20 in this recording

  5. Our readingAI is still terrible at last mile fit and finish.

    Their wordsThe frustrating thing is like AI is still terrible at last mile fit and finish.

    ↗Brian Lovin - How to level up with AI as a designeryoutube.com 7th of 20 in this recording

  7. Our readingThe models are still bad enough that if you really care about the final quality of what you are shipping, you probably have to look at the code.

    Their wordsI think the models are still bad enough that if you really care about the final quality of the thing that you're shipping, I think you got to look at code.

    ↗Brian Lovin - How to level up with AI as a designeryoutube.com 8th of 20 in this recording

  9. Our readingFeeling you have a grip on AI by getting your skills in the right order and your review flow right is probably a lot of cope for the fact that the models are still inconsistent.

    Their wordsbut feeling like you have a grip on it like oh if I just get my skills in the right order if I just review things in the right right flow and like have this model do this. I think it's a lot of cope for the fact that like the models are still inconsistent.

    ↗Brian Lovin - How to level up with AI as a designeryoutube.com 10th of 20 in this recording

  11. Our readingHigher quality input tokens give higher quality output tokens, and the input tokens are where you steer where the model goes.

    Their wordsThat means if you have really high quality input tokens, you're going to get higher quality output tokens but those input tokens are where you get to steer where you want the thing to go.

    ↗Brian Lovin - How to level up with AI as a designeryoutube.com 11th of 20 in this recording

  13. Our readingIf you want the model to pay attention to a certain thing, you have to say that thing.

    Their wordsif you want it to pay attention to a certain thing, you have to say the certain thing. Like you actually have to say the thing that you want it to focus on.

    ↗Brian Lovin - How to level up with AI as a designeryoutube.com 12th of 20 in this recording

  15. Our readingStupid prompts can work, because they flip the right bits so that the model will extrapolate from there.

    Their wordsThese are stupid these are stupid prompts but they work because it just flips the right bits so that the model will extrapolate from there.

    ↗Brian Lovin - How to level up with AI as a designeryoutube.com 13th of 20 in this recording

  17. Our readingAnyone who knew how to code before AI is having a way better time using AI than someone who did not, because they know the right words to say.

    Their wordsanyone who knew how to code before AI is having a way better time using AI than someone who didn't know how to code because guess what? They know the right words to say.

    ↗Brian Lovin - How to level up with AI as a designeryoutube.com 14th of 20 in this recording

  19. Our readingAI is not magic: the underlying systems are very complicated, but what it does on your computer, like writing files and invoking bash commands from the 70s, is knowable.

    Their wordsAI is not magic. The underlying systems are very complicated. I don't think anybody would claim they know what an LLM's doing under the hood. But it is doing things on your computer like writing files and writing scripts and like invoking bash commands that were invented in the 70s. And those are knowable things.

    ↗Brian Lovin - How to level up with AI as a designeryoutube.com 15th of 20 in this recording