We don’t need more AI

We are currently at a moment in time similar to the dawn of the internet: everyone is trying to figure out how to use AI and agents to build businesses and generate a profit. There are AI courses, workshops, and startups everywhere, and we’re trapped between "doom scenarios" and overly optimistic ones.

The truth is that not everything needs AI, although the technology can and has proved to be extremely efficient in many industries. We don’t need AI for the sake of AI, no matter how exciting the technology is or how strong the illusion of productivity is. 

Not every process needs to be automated, and not every business needs to add an extra layer of complexity on top of what they already do. Consider Craigslist, a website that is essentially stuck in time, yet works exactly as it is. 

Think of companies that operate in a traditional way where technology will never be able to fully replace manual processes. In real estate for example, people need and will always look for human connection. As a proptech founder, I actually advised a residential developer against using an AI solution. I didn't think it would work for the geography and for the place his business was in. It wouldn't have worked because, no matter how efficient or productive AI agents might be, we remain deeply emotional beings who seek to connect with others when making a big decision like buying our next house.

In fact, we have reached a paradox: we have so much access to information and data that we are actually looking for ways to escape the internet. We want to go back to in-person interactions and physical events.

Furthermore, AI has become more expensive to use. You can prompt AI to do so much that you actually end up overwhelmed by the possibilities. We are not working less, as the theory promised - we are working more. 

As tech founders, what does this mean for us? 

First, we have to acknowledge that not all business ideas need AI. Many models work perfectly well without this technology, or can use it to optimize some parts of the process, not disrupt the entire model.

Second, AI should be used only if it adds real value. We have to move beyond vague promises of "efficiency." What does your AI business really return to your customers in terms of quantitative and qualitative outputs? What is the actual problem it solves? If you cannot answer this, your solution will fail - in fact, roughly 95% of AI pilot projects fail. 

Similarly, I’ve seen cases where people thought they needed AI to be more efficient in generating leads, but the real problem was elsewhere - teams were actually not managing leads well enough. The prospects came, but they weren’t attended to. So it’s very important for both the customer to understand the problem and for us as founders to identify it correctly.

Third, I think that there’s real value in combining digital products and solutions with in-person connection. The community is the moat and big players have proved it - think of Apple, which has invested so much in culture and brand. This is a way for companies to stay relevant and build a legacy. When I started hosting in-person meetups, I was surprised by the big number of people who registered - so I had my proof of concept.

Truth is, we need to know when innovation should stop. We need to filter use cases and know what does not need AI. The world doesn’t need another digital solution that will make it even more isolated and disconnected - but one that adds real value and facilitates real connection.

AI is not the magical solution that will solve all our problems. We are.

Photo by Egor Komarov on Unsplash 

Podcast feature: Ginger Shot (RO Only)

Georgiana Vintila invited me to join her podcast, Ginger Shot! Check out the episode to hear more about expanding into an international market, navigating conflicts and cultural differences, and the internal transformation that the business growth brought.

Thank you for having me, Georgiana!

A few words about me

I’m Irina Constantin—entrepreneur, educator, and avid reader with over a decade of experience in the tech industry. I’m the CEO and co-founder of VAUNT, a sales and marketing OS for residential developers that powered $600M+ in 2025 transactions across 90+ developers.

I am a second-time proptech founder, I run In The Built World, a community connecting real estate operators with tech innovators, and host the From 0 to 100M podcast on what it actually takes to scale. Forbes 30 Under 30.

If this newsletter was forwarded to you, subscribe here so you’ll get the next one in your inbox.