> I haven't seen any real competitor of YouTube, or Figma, or Google, or Instagram created solely with AI.

Is a ridiculous set of products to expect competitors for. First all but 1 rely heavily on network effects. YouTube is a low cost provider. Think about it, they provide unlimited high quality video storage for free. They process 500 hours of content every minute. Google and Instagram have a huge network all users benefit from. AI can't help you bootstrap this network.

The only one that's fair is Figma, and considering Figma is down 30% from IPO price (much lower from ending day close), it's fair to say there are competitors to Figma, including the LLMs themselves.

That begs the question: what is product? Is "an app for designers to change style and communicate with developers" the product? Or is the product "to make websites look pretty and well organized"? The former, Figma is great. But if the product is the latter, LLMs are a real threat. No one would bother building a "better Figma" today, why would they? They would just solve the problem Figma purports to solve

Not in any academic sense, but in a real-world sense a software product is something that as a user you can rely on to solve a problem. Rely is doing the heavy lifting in that sentence.

You can rely on a vendor’s product if (a) they’re going to be around for a long time, otherwise the investment into adopting it isn’t worth it; (b) the vendor is committed to solving your problem in an ongoing way - meaning they understand enough of the domain (ideally more than you do) that they’re going to be a partner even as your own needs evolve; (c) they deliver the product in a dependable way - high uptime and security.

So it comes down to “can you depend on this software over time?”

The underlying point --that network effects exist and determine the success of a product-- has always been the case, long before AI made it possible to vibe code a rough Instagram clone in a weekend.

Anyway, if you want a successful new product, the path to success (obviously) isn't likely to be copying an existing behemoth, but offering something new, unique and valuable. That's always been the (exceptionally difficult) challenge, and AI doesn't help with this. But I think we can be confident that in the future, as humans do create new, unique, and valuable products, some of them will probably be vibe coded.

This misses the mark.

A product is something which can be exchanged with others, transferred into their agency, for them to own and use according to their free will.

They don't have to use it or indeed 'be a user' - it is up to them whether or not utility is derived. The only requirement is that it is exchangeable and that agency has been transferred in such a manner that all parties involved in the exchange are satisfied with the transfer of agency.

A project is some activity which has an expectation of changing state over time.

Project management means projecting the scope and sphere of that activity into the future, in order to change its state in such a way, hopefully, that something exchangeable is produced. Product management means, making sure a satisfactory exchange occurs between the creator of a product and its new owner.

So many people get this wrong.

It is not a product if it's a project. It is only a product once it has been exchanged with another entity for their purposes.

You can have many, many projects which will never become products, because they won't be exchanged with others. You can have products which are valued by their owner in ways unintended by their creators. The value and utility is arbitrary to whether or not something is a product - the only requirement for something to become a product is that it is exchanged.

Successful product(S) are exchanged at scale. A single product can be very successful if the scale of that exchange is high. A lot of products can be very successful if the number of exchanges is great.

Until the exchange occurs, it's not a product. It's a resource. Inventory.

Side projects become products only once they are exchanged.

There really are plenty of Google analogs. There are plenty of chatbot analogs. Any one of us can recreate Instagram or X. These competitive analogs simply have not become products, because they have not been exchanged. Products which harness human collectivity have the problem of not being exchanged at scale until the collective decides to participate in the exchange, i.e. recommend their friends sign up to join them.

The problem with easily started new side projects is that they are not so easily exchanged at sufficient scale.

It is not that we didn't have the expertise to do so to begin with, but it would simply have been too expensive and time consuming. That was solved once the models became good enough.

There's a lot of money in the SaaS world, where the products are both unglamorous and one-trick pony-ish. Making your product LLM-proof can be challenging today, unless you happen to own a lot of data or lots of compute, and those two are the drivers behind your product.

Another mistake everyone making is not building better and bigger products. What is used to take a month to develop is now possible in few days but most don't put the same effort as before to make bigger and better products. They just do they same as what they used to do before but quickly. This is a mistake. AI uplifted everything so products built with AI should go to next level.

I'm an indie mac app developer. My first product took me 6 months to develop without much AI. I just released my second app ( https://www.vinaa.ai/ ) but took the same time to develop as I wanted it to be the best AI chat app. So I spent a month building my own markdown renderer in TextKit as existing library are not fast enough. I wouldn't done this if I don't use AI.

The vibe-coded demos everyone's tired of are code with no environmental need pulling on them.

So building is faster, but marketing—not so much

But the question of should I spin up my own website in a few minutes or use wordpress for my blog is now open. Your personal tools can compete with what's out there and you may decide what you build in one session is more compelling than what the "pros" are offering.

And for personal tools you are much more forgiving for the mistakes if it gets whatever task you want done.

E.g. it's now about the delivery of a service rather than the challenge of building it (e.g. the moat) and then the delivery of service of it afterwards. Non functional requirements like data management and security, legal requirements and hosting are still valuable things that aren't front of mind for most developers creating their own tooling.

Imo a better distinguisher of demos and products is: "is your thing CAPABLE of solving a real problem" (even if it isn't doing so at a scale atm). Google+ was a product (not a super successfull one), eventhough it didn't have many users. Conversely, a viral interactive gimmick can attract millions of users in a day without really being a product at all.

AI obviously is powerful and quick to build, but intellect, intelligence, and creativity will still lead to disproportional outcomes for users. After all, AI still relies on effective prompting to build well.

The real danger, then, is not AI churning out endless slop, it's actually that the slop degrades the perceived quality of actual products and legitimate effort.

Measuring by users is one metric, yes, and a good one. But, quality of output and performance beyond a simple prompt is what will make the difference. Products couldn't only be measured by the number of users, but also by quality and ingenuity.

Lastly, the products of tomorrow will not defy AI, but will build on top of them. AI will build and perform a higher level of compute, yes, but the products will orchestrate AI workflows, rather than deterministic outputs. Guided reasoning, orchestration, and value creation are human qualities, not AI.

The word ‘product just left everyone confused.

If one uses a word like platform, application, prototype, offering, ‘products’ this makes it clearer than just the generic ‘product’. It’s even better if goal is applied as well. For example Flight finder Application, or Video Playback Platform.