Paul Graham resurfaces his 2012 essay arguing Y Combinator application trends serve as early indicators of major technology shifts

@paulg What does the future hold based on current applications?

@RealDayneTran The fascinating thing is that this question is still unanswered. And yet we know there is an answer, and it will seem obvious in retrospect.

@plowingthedark Pervasive, transformative uses of AI.

I just came across this essay I wrote in 2012 predicting that hard tech startups would become a big thing. Not for the last time, YC applications turned out to be a good predictor of future trends.

The Hardware Renaissance: https://paulgraham.com/hw.html

@paulg what hardware form do you like the most for ai? glasses, pendant, still a phone?

@paulg The accuracy of this is incredible. The emergence of affordable, powerful microcontrollers (like ESP32) and long-range protocols (like LoRa) completely democratized the barrier to entry. Hardware prototyping shifted from an enterprise luxury to a weekend software extension.

@paulg My strategy analysis !

⬇️as follows 👇 👇 👇

@paulg Personal hardware seems compelling for the first time. If CAD can give way to natural language hardware design people (me) might consider buying 3d printers

@paulg Hardware got a shit ton easier in the past 6 months.

@paulg Odds don't match up but gut feeling AI will benefit humanity at large.

@paulg PG: *makes basic, predictable observation* PG Fanboys: *go absolutely berserk*

@paulg To paraphrase: “people who really love software end up making their own hardware.”

I think that was Kay, or Jobs quoting Kay.

@paulg Were easy tech startups a thing before 2012?

I missed my windows to be a founder and buy Bitcoin at the same time...

@paulg @ycombinator I think there are two main reasons behind this: - Building hardware has become much cheaper and easier over the years; - There are now many successful hard tech startups, which motivate more people to start their own.

At least these were some of the reasons we started ours!

@paulg It looks like your shortest essay ever

@paulg @DanielleFong I want to build eyeballs for Rosie Chatgippity

@paulg This essay was spot on, the shift is real.

@paulg Right about the what, early on the when. What changed wasn't the appetite for hard tech — it's that tooling got cheap enough that a small team could attempt atoms, not just bits. The prediction was correct; the unlock was the cost curve.

@paulg Bro, nothing ages better than “I predicted this” posts that are written after it already happened.

@paulg Funny how YC apps have been quietly mapping the future for over a decade. This essay still feels fresh.