AI has its limits

getty

I recently had an opportunity to serve as judge looking at several innovative startups, and was impressed with how these entrepreneurs are spinning up new businesses with artificial intelligence. While some had developed products to better manage AI and AI agents, a couple had broader uses.

Gustav Hugod, head of sales and user experience at Atech, for example, demonstrated his company’s kit that included some small hardware components that enabled users to design prototypes of machines or systems – such as robots or weather stations – with verbal prompts at the front end.

Esther Sauri, marketing director at another startup called Gataca, promoted a mobile interface that accelerated and expanded the range of digital wallets to include documents as various as passports, driver’s licenses, and even college transcripts. Gataca is positioning its offerings to capitalize on the EU’s upcoming mandate on making documents available on digital wallets.

What’s the secret to succeeding these days as a tech-driven company? In one famous and timeless give-and-take, Steve Jobs said it best: “Start with the customer experience and work backwards to the technology. You can’t start with the technology and figure out where you’re going to try to sell it.”

The two companies mentioned above are examples of looking at what customers are seeking, and working backwards to build technology to meet those needs.

Remember, you can hand the latest, most sophisticated film-making equipment to someone – infinite film-making power – but that’s not going to make them the next Steven Spielberg. It takes that extra human factor, the sense of what people want and value, to deliver.

AI isn’t the end-all, but can be a vital tool for achieving this. “What truly resonates with customers is not only speed or convenience; it’s the feeling of being understood” write Brian Bischoff and Bree Basham, both with CapTech, in Harvard Business Review. “AI is transforming today’s customer experience by creating interactions that feel less transactional and more relational. Intelligent systems can detect frustration, urgency, or delight in real time, allowing brands to respond with empathy and reassurance.”

They refer to AI assisting in the rise of what they call the “prototype economy,” with rapid prototyping and accelerated product development cycles.

Such is the root of innovation. And still very much a subject of debate, as the world drowns itself in AI hype and speculation. AI has the ability to spur new innovations, and we’re starting to see examples blooming across the world. There’s also an important lesson here as well: AI itself does not come up with innovations, and does not figure out what customers want. It takes people to do that. It will always take people to do that.

David Lieb, partner at Y Combinator as well as co-founder and former CEO of Bump, posited this thought experiment in a recent post on X/Twitter: “If every company suddenly had infinite free compute, what new products would emerge? My take: with very few exceptions, not much would change. The bottleneck is figuring out what people want, and it’s not so easy to apply compute to solve that.”

Lieb’s post drew many interesting responses. "Building a real product takes much more than just crunching numbers or having endless computing power,” responded Aditya Patro, founder of Inference. “Sure, you can iterate faster and test ideas more quickly, but that doesn’t magically reveal what people truly need or will stick with day after day. It’s all the human factors that get in the way – understanding the messy problems folks face, perfecting the right feel and simplicity, and spreading the word without it feeling forced. Taste matters a lot here. Even with all the tools available, most attempts end up as cluttered messes or projects nobody asked for?”