While there are many trending and debate-worthy topics within the world of artificial intelligence, theres a rather profound one that few people are starting to discuss openly. Namely, theres a fundamental disconnect between how the tech industry communicates about innovations in AI versus the actual value delivered to consumers and enterprises. Worse, AI is not fully democratized and is not the bastion of major tech companies. Fortunately, that is about to change. <\/p>\n
As consumers, we are awash in a myriad of daily stories concerning AI and machine learning, from IBM Watsons latest use case to the warnings of Stephen Hawking 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 AIs potential to impact industries on a global scale, such as in health care, advertising, finance, security, and more. <\/p>\n
The impression we have, therefore, is that AI is widespread and easily accessible that humanity is benefiting from these groundbreaking applications, and that they impact our lives in a meaningful and helpful way. But this is a big misconception. The reality is that we have enormous strides to make in democratizing AI development, and making these innovations truly accessible and available to mankind. <\/p>\n
As it stands now, the vast majority of AI is being developed within the enormous black hole of a few major technology platform companies. These household name tech giants are monopolizing the best and brightest human capital, and they have access to Big Data and other critical resources which is limiting the ability of other major global enterprises, let alone small to mid-size companies, to compete. These industry giants have their own specific business models and requirements, and as a result, they tend to focus on a relatively limited subset of AI applications. The problems they are tackling, while very real and worthwhile, are still just a tiny portion of AIs potential to impact specific industry vertical use cases and the overall economy, not to mention humanity as a whole. <\/p>\n
A few tech industry titans thus control the vast majority of talent, data, and other resources necessary to develop life-changing technologies, and this is bad for any number of stakeholders who stand to benefit from AI. Competition should happen at the application and business levels, and not be based on a single industry monopoly. <\/p>\n
The good news is that weve reached a tipping point, and AI is actually helping to shift the dynamics and level the playing field. As our systems become more advanced and the costs to develop new AI software begin their predictable fall, its becoming easier and easier for startups and smaller companies to rise up against the tech giants. Rather than focusing on a confined set of problems, these up-and-coming players will be free to cook up innovative, disruptive solutions that arent restricted by existing business models and product services. <\/p>\n
Consider the infamous Innovators Dilemma as demonstrated by Clayton Christensen. If youre not familiar, the idea is that successful companies (so-called incumbents) can do everything right and by the books, yet theyll still lose their market leadership to new and rising competitors. There are two key parts to this dilemma. One is that the value to innovation is an S-curve, meaning that product improvement necessarily takes time and involves multiple iterations. By finding the right application and market, startups are able to find the sweet spot of value using iteration at a much faster rate, and thus enter and disrupt the more mature markets of the incumbents. <\/p>\n
The second is the idea of incumbent-sized deals which means, while incumbents may have the advantage of a huge customer base, this carries higher expectations for yearly sales and performance. Startups dont need to worry as much about these requirements, and thus have more time and energy to focus on innovating a new entry, next-gen product. <\/p>\n
AI has so many applications beyond the business needs of a few of black-hole tech platforms. Weve reached an exciting time when emerging technologies are facilitating smarter, faster, and better businesses processes at increasingly lower costs, and this 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 had been ignored by their past employers. As more newcomers and startups make progress in AI development, we will surely witness a broader spectrum of adoption and, thus, a much greater and meaningful impact on society. <\/p>\n
Roger Jin is the cofounder and CEO ofRul.ai,focusing on AI technologies. <\/p>\n
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. <\/p>\n
<\/p>\n
Read this article: <\/p>\n
The dangers of letting Big Tech control AI - VentureBeat<\/a><\/p>\n","protected":false},"excerpt":{"rendered":" While there are many trending and debate-worthy topics within the world of artificial intelligence, theres a rather profound one that few people are starting to discuss openly. Namely, theres a fundamental disconnect between how the tech industry communicates about innovations in AI versus the actual value delivered to consumers and enterprises. Worse, AI is not fully democratized and is not the bastion of major tech companies. Continue reading