A bet on the future

We are building Alter, a native app that brings AI capabilities to all your apps on Mac. We choose to build this product on the Apple platform, this post explains why.

Apple has two unique features that make it standout when you want to build an AI product:

  1. A unified memory.

    AI models need a lot of memory to work, at this point most of the useful models are hosted in the cloud. It's too memory and compute intensive to be useful on a personal computer.

    But this might change in the future, the quality and speed of the models are scaling. If it keeps scaling on both sides, we might soon have the option to run models on our computers which open interesting properties for building apps: privacy-first, low-cost and speed (as you remove network latencies).

  2. Apple Script

    Yes, it's old. Yes, it's slow. But it's there. It's like having a secret door to control apps that Windows or Linux users can only dream of. It's not about the speed; it's about the possibility.

    I'm not even sure if Apple is aware of how fantastic it is in the context of AI agents.

Fast-forward to the future, mix these two ingredients and put entrepreneurs with a creative vision of what AI could hold at the OS level and…

We might not be far from HER.

A unique experience where you have control yet an invisible friend helping you to achieve more with the tools you know. You don't need to learn something new, you just have to remember it's there. And you will, because it's seamless and always there.

Of course, we are at the first steps from a long walk. With Alter we're focusing to get these first steps right.

Yet the exciting part is elsewhere.

Apple is lagging.

While Microsoft has its Copilot 365, Apple hasn't yet shown its hand in AI. It's a huge opportunity for a small startup to lead what AI on macOS means.

And when Apple does step into AI, they will propose their own vision of it.

It's another chance to stand out, a chance to be part of the discussions triggered by the Apple launch.

Being there first is a unique mover advantage, and Apple might take longer than we expect. iOS will be served first, but who knows when for macOS?

Also, by supporting macOS down to version 13, we're not just preparing for the future; we're making sure everyone can join. It's a vision for a smart world that includes everyone. Not sure that Apple will take care as we do on this side.

So, what might look like a thoughtless choice is, in reality, a careful one.

It's also bold.

When you know that a company like Apple is going to market and embed features similar to your product in less than 3 years, you think twice before committing. But sometimes, you have to look at the cracks, try to exploit them, and bet that it will pay off.

With Alter, we believe we can be fast enough to bring a unique and distinctive vision of what AI can do at the OS level.

Let's place our bet!

- Sam

Reply

or to participate.