Ask HN: Is it worth building a non-AI software company?
I have a product idea around cloud security and I have been working on it for the past few months. Quit my job about 4 months ago to work on it full time.
But lately, I am getting distracted because I keep wondering that whatever I am building can be built by anyone with the help of AI. Moreover, I see big companies launching AI models focusses around cyber security. for ex: the sec-gemini model released yesterday by google.
I can't shake the feeling that I am not building anything novel and someone with basic software engineering background can replicate the idea with the help of AI.
what are your thoughts on it?
I'm doing a similar thing (not could security though), similar thoughts try to surface.
If you're trying to build a business, only a small part of it is about generating code. Sticking with it and developing a good product is where you make the difference. I've built many things in a couple of weeks of coding, but then given up when faced with the hard work of making it a solid business.
My other thought is.. There are always others coding faster than me, writing better web copy, doing sales more successfully, etc. You/we pick AI as a threat because we understand it, but a good sales person is just as much of a threat to your idea.
We all have technical/skills advantages and disadvantages. In the successful startups I worked in these were not that important in the end. It was just about keeping going, every day. Relentless optimisim that you're right.
You/we pick AI as a threat because we understand it, but a good sales person is just as much of a threat to your idea.
this is a great insight. never thought of it that way. thanks so much
I think it's better to focus on the problem.
In one of the YC videos I think they called that a "solution in search of a problem". A couple years ago everything was blockchain technology searching for a problem, then it was VR technology searching for a problem, and before that it was social apps... but if you start with a problem first it's easier to build the right solution and skip the latest hype cycle.
I think AI is hyped up a lot right now and there's definitely going to be a trough of disillusionment next or another AI winter, especially as all these AI startups fail to find a successful business model.
If you are worried about someone replicating the idea with AI easily then you can spend some time focusing on your business moat, which AI companies are struggling with now, but especially with cloud security or cyber security I would be worried about trusting anything that was vibe coded or had AI integrated because it's a black box.
> I can't shake the feeling that I am not building anything novel and someone with basic software engineering background can replicate the idea with the help of AI.
Why are you so fussed about what tool is being used? The product should be novel, not whether it's AI or not.
As long as the product is good, why would customers care if it's AI or not? There have also been plenty of existing products calling themselves AI if that's all you're after.
The world is huge. Even without AI, there are enough people that can replicate most things these days. OpenAI thought it was "safe". No so eh?
but that's the thing. Its not novel. As in, I am inventing something. I see a gap and opportunity and began to execute.
I am worried about AI because it has reduced the entry barriers. Much easier now to write code or atleast build MVP and get traction. and that makes me question whether all this effort is futile if the only things giving me an edge os the early start.