what if ai had taste

i’ve been having a lot of conversations with people building in the ai space. and it got me really excited to explore something myself hahaha.

so here’s what i find interesting.

i run a startup, and i found myself spending an awful lot of time trying to get high quality customers, attract top talent, and raise from tier-1 vcs.

the moment we sort of figured out how to do this on a small scale, we realised how much of it came down to distribution.

and that got me thinking about what ai could help us do here.

but before i go into that, let’s start with what distribution actually is.

distribution is how your product actually gets into people’s hands, repeatedly, at scale.

we’ve hit a point with today’s technology where you can reach people almost anywhere on the planet. but reaching them is only part of the problem.

just to really let this sit with you reading this.

a long time ago, people would stand in the town square and yell. if you wanted people to hear what you were selling, you had to make yourself heard.

many years later, when printing and newspapers were introduced, people could rent a tiny box on a page that a big portion of the people living there read. you could pay to get in front of an audience.

we got radio and television. a handful of channels that had massive audiences. if you could afford a spot, you could reach millions. just being able to be there was already a huge advantage.

the internet. yes, this really opened the gates. websites, search, social. the cost to publish something dropped to almost zero, and the number of places you could publish became enormous.

now we are here. publishing is basically free, and anyone can potentially reach millions.

but a lot of reach can still lead to very little.

you have to make something people actually stop for, consume, remember, and act on.

and that feels harder than ever.

in 2014, peter thiel wrote about this in zero to one. how companies can fail because they haven’t figured out how to sell what they’ve built.

this was twelve years ago, and it feels even more relevant now, with a larger weight on the actual message.

so this is where i get interested in ai.

i believe we are at a point where we can scale taste if we can actually deconstruct it.

meaning, what if we could take “a diamond is forever”, “think different”, “red bull gives you wings”, “finger lickin’ good”, and all this other messaging we will never forget, and understand what made it work?

then feed that understanding into the ai for your company, and have it help your team distribute what it is you are selling.

help the right customers understand why they need it. help the people you want to hire understand why they would want to build it with you. help investors understand why it could become something big.

this way, we could allow teams to spend more of their time actually solving the problem their customers face.

and we can do the same for ours.

because i believe poor distribution, not product, is still the #1 cause of startup failure.