Is there an AI bubble? From the head of Google’s parent company to officials from the Bank of England, many fear that the AI’s growth is being driven by massive investment and hype, similar to the dot-com boom.

As it stands today, the unit economics of the AI industry don’t add up. The revenue or value an AI company or a large language model like ChatGPT earns from servicing a single user or query is less than the amount it earns from that interaction. OpenAI, for example, is not profitable, even though it generates billions of dollars in annual revenue. Spending on operating costs, compute, research, and development far outpace revenue.

Yet, investors are still pouring billions into this technology. About $202 billion was invested globally in AI in 2025, according to Crunchbase. This includes AI infrastructure, foundation labs, and applications. AI was also the leading sector for startup funding globally from 2023 to 2025. 

In the MENA region, AI startups raised over $2 billion in the first half of 2025, a 134 percent year-on-year increase, according to HUB71.

IMPACT ON FUTURE INVESTMENT

For some founders in the region, the fear of a potential overhype and overvaluation is concerning. “Even if you’re looking to build applications that are commercially viable, if there is big noise around you, this will impact everyone’s decision-making,” says Ahmed Mahmoud, CEO of DXwand, a startup that leverages AI to automate customer service and employee assistance. “It could impact any future investment.”

It’s not only startups that have invested heavily in AI, but also enterprises and corporations. Many have allocated significant resources to Graphics Processing Units (GPUs), servers, and subscription services, but some, if not all, are still waiting to see returns on their investment. “This will make all decision makers think if they’ve done the right thing,” says Mahmoud. 

To ensure AI delivers real value, he offers SMEs AI-powered sales agents that help acquire and convert customers at a fraction of the cost of hiring human sales staff.

OPPORTUNITIES FOR STARTUPS

Mohamed Hussain, Principal at Anara Capital, while he recognizes the hype, overvaluation, and risks associated with the underlying technology, sees real demand for it. One that founders can leverage if they know how to use it right. 

“There’s definitely huge capital inflows into AI companies, inflated valuations, and a lot of AI washing,” he says. “A lot of AI is being used for marketing purposes rather than for actual operating purposes. So yes, I do think there’s definitely some hype. And I do think there has to be a market correction at some point, which will result in many AI startups disappearing over the next few years, but I also think there is huge, tangible demand for AI.”

He adds: “At the foundational level, like data centers and compute platforms, we’re seeing millions and millions of users. There’s real demand, serious money, and serious traction.”

The key, according to Hussain, is for startups to use AI to solve real problems rather than as a tool to attract funding, especially to streamline operations and increase efficiency. He gives an example of a fintech startup assessing credit scores. By using AI, the startup can reduce processing time, increase efficiency, and perhaps even cut costs by becoming leaner. 

“AI is now becoming an expectation, at least in the tech startups world,” he says. “So, we always have to look one step further. How are they using AI? Is their AI different, or are they simply integrating it with ChatGPT at the back end? So, we really assess what AI means to them. And we also assess what impact AI is having on the operations of companies.”

In addition to increased efficiency, AI can also improve a startup’s economic margins, making its unit economics more sustainable and putting it on the road to profitability sooner rather than later. All this gives founders the opportunity to improve their products and services, and reach new markets quickly.

“There are definitely expectations on startups,” says Hussain. “They’ll need to make sure that their business and solution are defensible and scale that much quicker and take advantage of the capital inflows for now because that’s not going to be forever. But at the same time, there’s also a lot of opportunities for them across the board.”

RISK OF UNDERINVESTING

To keep innovating and creating value, however, founders need money. 

Akshat Prakash, CTO at CAMB.AI, a startup that live streams in multiple languages, says there is a risk in underinvesting rather than overinvesting in AI. “It’s not debatable that AI is going to be consequential,” he adds. “We’re not debating here whether AI is going to be the fundamental backbone of any vertical. People already believe that. The question is, is it going to be in 10, 15, or 20 years?”

He says that the mindset of investing in AI needs to change. Unlike other technologies, AI must be deployed and invested in before it generates returns. “We do live streaming for the world’s biggest leagues in multiple languages,” he says. “This technology will never be complete. It can only get better and better. So, it must be bought first and then scaled. So, enterprises and companies are just not used to buying something that doesn’t work yet.”

Prakash likens it to companies like Google and Facebook when they first started. In their early days, they weren’t profitable. They focused on creating value first, in getting people to use their platforms before becoming some of the world’s biggest companies. “These companies made a profit later,” he explains. “New technologies like AI must be deployed before they can be fully monetized; those profit cycles are even longer.” 

If startups can’t see that, they are missing the bigger picture, he adds.