Among the many trending and debate-worthy topics within the world of artificial intelligence, a few with rather profound ramifications are finally being openly discussed. Namely, there’s a fundamental disconnect between the way the tech industry talks about innovations in AI and the actual value delivered to consumers and enterprises. Worse still, AI is not yet fully democratized and has remained largely the bastion of major tech companies. Fortunately, that is about to change.

As consumers, we are awash in daily stories concerning AI and machine learning, from IBM Watson’s latest use case to Stephen Hawking’s warnings on the subject to the rise of AI-style terminators. The average user is perhaps vaguely aware that AI powers everything from their inbox to their music playlists to their social media feeds. Savvier users may be more familiar with AI’s massive potential to impact such industries as health care, advertising, finance, security, and more.

But the impression we have is that AI is widespread and easily accessible — that we all benefit alike from these groundbreaking applications. This is a big misconception. The reality is that we have an enormous way to go if we are to make AI truly accessible and take advantage of its myriad potential use cases.

As it stands now, the vast majority of AI is being developed within the enormous black hole of a few major technology companies. These industry giants are monopolizing the best and brightest human capital and they have outsized access when it comes to Big Data and other critical resources. This naturally limits the ability of many global enterprises, let alone small and mid-size companies, to compete. As the industry giants have their own specific agendas, they tend to focus on a relatively limited subset of AI applications. The problems they are tackling, even if real and worthwhile, address just a tiny portion of AI’s potential to impact the tech industry and the overall economy, not to mention humanity as a whole.

These companies’ control of the vast majority of talent, data, and other resources necessary to develop life-changing technologies is bad for any number of stakeholders who would otherwise stand to benefit from AI. To foster greater innovation and inclusion, competition needs to happen at the application and business levels.

The good news is that we’ve reached a tipping point and AI is actually helping shift the dynamics and level the playing field. As our systems become more advanced and the costs of developing new AI software begin their predictable descent, it’s becoming easier for startups and smaller companies to get in on the action. Rather than focusing on a confined set of problems, these up-and-coming players will be free to cook up innovative, disruptive solutions that aren’t restricted by existing business models and product services.

Consider the thesis of Clayton Christensen Innovator’s Dilemma. If you’re not familiar with the book, the idea is that successful companies (“incumbents”) can do everything right but they’ll still lose their market lead to new and rising competitors. There are two key elements to this theory. One is that innovation happens along an S-curve, meaning that product improvement necessarily takes time and involves multiple iterations. In searching for the right application and market, startups are now able to find their sweet spot by iterating at a much faster rate and can thus enter and disrupt more mature markets dominated by the incumbents.

The second idea addresses “incumbent-sized deals,” which means that while incumbents may have the advantage of a huge customer base, they also carry higher expectations in terms of yearly sales and performance. Startups don’t need to worry as much about these requirements and thus have more time and energy to focus on innovating a next-gen product.

AI certainly has many applications beyond the business needs of a few of black-hole tech platforms. We’ve reached an exciting time when emerging technologies are facilitating smarter, faster, and better processes at increasingly lower costs, which is opening up the playing field to smaller, leaner players. It will become more and more common to see five-person startups go up against the tech behemoths. Top AI talent that has been incubated inside these companies will inevitably start to leave and create their own startups, addressing new use cases that were ignored by their previous employers. As more newcomers make progress in AI development, we will surely witness a broader spectrum of adoption, allowing AI to have a much greater and more meaningful impact on society.

Roger Jin is the cofounder and CEO of Rul.ai, focusing on AI technologies.

Above: The Machine Intelligence Landscape This article is part of our Artificial Intelligence series. You can download a high-resolution version of the landscape featuring 288 companies by clicking the image.