Is SaaS dead? What’s next when AI takes over?
I’m starting to think most SaaS products are on borrowed time. With AI getting smarter every day, it’s poised to swallow up their use cases. AI will learn to do what SaaS tools do and at some point it will be easier for the user just to use the AI. I recently saw a tweet claiming that by next year, 90% of code will be AI-generated. If development costs are plummeting to zero, why would anyone pay for a traditional SaaS subscription, when his favourite AI can do the same?
Do you see it the same way? I’d love to hear your take.
I love building SaaS products, but if AI is about to render them obsolete, what’s the point? So, if SaaS is dying, where should we redirect our energy? Custom AI solutions? Niche tools that AI can’t touch (yet)? Something entirely new? I’m curious, what would you focus on in a world where AI owns the software game?
Replies
My-legacy.ai
SaaS isn’t dying, it’s evolving. AI won’t replace it—it’ll just make the best products smarter. The real winners will be those who build AI-first, adaptive solutions. Curious to hear how others see this shift!
@mylegacy SaaS isnt dying just evolving faster with help of AI. Those who adopt faster will see the benefits at a longer run, IMHO
Graphify
What do you mean by “people won’t use software because AI can do the same”? Just typing a prompt into ChatGPT isn’t the same as using a well-designed SaaS product with a structured workflow, right?
Sure, AI is making things easier, but will people really start building everything from scratch just to avoid paying for a tool that already does the job better and faster? Most users don’t want to tinker with AI models, they just want a solution that works out of the box. Curious to hear your take, do you think AI will fully replace SaaS, or will it just shift how we interact with software?
@hussein_r I agree that just typing a prompt into AI isn't the same as using Saas that is purposfully build for a specific task.
However with the current trend and looking into the future I think AI will be able to do almost everything current Saas does. And I don't mean people will build alternative to any Saas just to avoid it. AI will be able to automate everything that is software related in my opinion. So people will just use the AI as it will be more convinient for them.
I think it will replace Saas AND it will shift haw we interact with software. With an advanced AI you wont need much more then a micrphone (for input) and a screen (for feedback) to do anything
Graphify
@mzkvisuals Interesting to hear, in what time frame do you think, this will happen? 🤔
@hussein_r probably 5 years but this is highly speculative
@hussein_r agree -- Enterprise tech solutions ("software" and otherwise) aren't going anywhere. AI adoption in the mainstream is still in its infancy -- something like 90%+ of companies in the US are SMBs (say <$500M annual rev). Who will help them adopt AI into their ops? SaaS providers.
@hussein_r @leah_madden True but for how long? Won't the AI at some point replace them all together?
Build SaaS that has inertia that cant be easily made even when AI gets better. The challenge is of course that inertia is the difficult part.
@sentry_co I absolutely agree :) But look at what is happening now. Big SaaS companies are being replaced by solopreneurs or small AI powered teams. The next step is full automation, AI replacing those small teams.
Yes. AI is a huge oppertunity, but its not like earlier tech opportunities, internet, crypto, etc. Because AI in it self is flattening the moat you previously built up when you made a product position within these tech waves. So this time, you have to think about things a bit different. If you already know the tech moat will be weak, even if you find PMF. Then the emphasis must be on other things. Hardware, Branding, Product position, Monopolies, complexity, Niche, bespoke solutions. Frontier tech. Funding. There is a lot more as well. As lng as humans will have problems, there will be need for solutions. I would not be so alarmed by Edge lords on X, screaming about its all all over. These voices only say things that will give them attention, they usually are not rooted in reallity but are made to bate people to participate and generate view counts. Try to ignore edge lord posts. And keep an objective mindset towards the future. Things will change. But at the same time, there never has been a more excited time to be a maker. The challange is now the imagination, and to use time well when executing on that imagination.
@sentry_co I agree. Right now the focus is shifting from building to other areas (branding, sales...). Tech companies will look more and more like media/marketing company then a dev company. But in the next years even if you have better branding, if ChatGPT does the same thing as your Saas, people will slowly shift to AI, more convinient and included in what you already pay for.
@sentry_co @mzkvisuals let's see how long ChatGPT lasts. with all other options on the market offered by respected and well established companies it's not a certainty
@sentry_co @taniabell Yeah I wrote ChatGPT but I ment just any bigger AI leader (grok, google, deepseer...)
I think that the next wave will be AI-first products that think, adapt, and anticipate user needs better than ever. The question isn’t if SaaS will survive, but what it will become.
@erkin_gonultas Interesting thinking. I think this wave might be just starting now with MCPs
I feel like It's at a point where developers really need to start being creative and adapt with the AI to use it in ways that not everyone else is even if it's a small tweak it really makes a difference and easy to stand out from all the rest. But SaaS will never be dead. I hope.
@gerome24 Yeah I don't want SaaS dead as well, after all it's what I do as well :) But still think it's heading to that direction. I totally agree that devs must get creative and adapt and I know we will see some awesome usecases in the future. (I just watched a video of Claude using MCP to control Blender, fuckin awedsome)
SaaS is thriving even more. AI enhances a SaaS product by adding extra functionality on top. If the only competitive advantage a startup has is an AI-dependent feature, then not only is it easier for competitors to build a similar product, but the chances of the startup securing funding from venture capitalists are also quite slim.
Check this video on what VC's think about this, lot of useful insights by Rajeev Dham from Sapphire Ventures:
I think AI is really overblown currently. It's giant bubble. What makes you think AI will take over? I see very little value in the current AI landscape. We see big tech companies trying to build stuff with AI but it's just not interesting or useful most of the time. AI is still terrible at so much and can't be trusted in many applications. If old people can teach you anything it's that they are much behind the tech curve so even if AI does do more old people will not use it and default to the stuff you know. My point is there will always be a market for non-ai products. So in the short term, barring some massive leap in AI technology, I don't even worry about it.