Does AI have a monopoly problem?

As captivating as AI’s rise into the limelight is, it’s also equally as concerning. The impact on the global economy is apparent and the ripple effect is something all industries are dealing with. 

However, there is another problem with AI that not many people are talking about - its centralized nature that could soon transform into a monopoly. 

‍

When tied to a location, centralized systems work well. There’s a catch, though. The internet has led to lightning-speed globalization, but in reality, our society is directly affected by centralized systems. Just look at social media and you’ll understand what we’re talking about.

‍

Undoubtedly, AI has staying power as it’s already disrupted the status quo in such a short time. But what happens when you put all that power into the hands of a select few privately-owned platforms? Platforms that are already dominating the landscape? 

‍

You get a monopoly that is impossible to break due to the complex nature of the underlying technology.

‍

What is an artificial monopoly?

‍

Before we address the intricacies of monopolies in terms of artificial intelligence, it’s important to get the basic definitions out of the way. 

Monopoly can be described as a market without competition - i.e. circumstances where an enterprise or an individual is the only supplier of a certain good. The main characteristics of monopolies are a lack of competition and substitute goods, as well as the potential for bloated prices that exceed the provider’s marginal cost. 

‍

Put all this into a blender and you get a high-profit margin and a situation where an entity has the power to exclude competitors - completely dominating the market in the process. 

‍

It’s important to stress that this market structure is not limited to large businesses. In some circumstances, smaller entities may be able to raise the pricing in niche industries or markets.

‍

But what’s the connection to AI?

Here’s a quick definition of an artificial monopoly:

‍

It’s a situation where an entity takes over the market due to factors such as advanced technology, regulatory advancements, or strategic decision-making.

Considering that the AI industry is dominated by OpenAI and Google, it’s easy to see how the current landscape is very close to turning into a bona fide artificial monopoly. 

‍

Reality of monopolization in the AI industry

‍

Artificial intelligence is a technology that allows machines to simulate human problem-solving capabilities. They work through models that analyze patterns in data sets. The entire system operates by receiving data inputs and reaching conclusions or taking actions according to the same conclusions. 

‍

Apart from the model itself, AI also relies on the hardware required to carry out requisite computations and train the model until it can deliver responses from queries or sets of queries. Another important aspect of an AI model is the data used for training it to perform the task itself. 

‍

Hardware required is expensive and developing new models takes a sizable amount of resources. At the same time, you can’t discount the data that it takes to support such a rapid development of AI and its overall operational efficiency. 

‍

Right now, the key players in the AI race are OpenAI and Google (with Apple as the latest contender) - all holding a significant amount of sway in the industry. Considering the blistering pace of innovation in the field, it far exceeds Moore’s law, leaving smaller competitors no chance of catching up. 

‍

It’s not just the hardware, though.

‍

Both organizations leading the pack have access to massive amounts of data they primarily receive for free from users using their AI solutions. As such, they are so much ahead of the curve that they’ll eventually end up running laps around their competitors, eliminating their fighting chance, and potentially leading to a monopoly (or duopoly or perhaps triopoly, to be more precise). 

‍

Dangers of AI monopolization

‍

Large generative AI companies have all the incentive in the world to vertically integrate with hardware providers. Since there are no regulatory standards or antitrust rules, the concentrated market is en route to causing stark inequality. 

For instance, the computational power required to come up with products in the market is immense - it cost Google over $190M to develop Gemini. Nvidia, which provides the hardware required to train AI models, has successfully capitalized on this, making it possible for them to raise the barrier of entry so high only the largest tech companies will be able to invest in AI technology. 

‍

It’s equally grim in terms of data required for training. Freely available data is slowly running out and many websites are blocking AI companies from utilizing their content for purposes of training AI models - completely locking out late entrants into the industry. 

‍

Since tech giants have a disproportionate amount of power in the internet tech stack, they have the means to control the nature, speed, and future of innovation. 

‍

By extension, such companies may also end up in a position to spread misinformation and propaganda, all the while suppressing trustworthy information. Put differently, the repercussions of AI monopolization may also seep into other areas of life.

‍

Have we reached the point of no return?

‍

There is still time to course correct. It won’t be easy as smaller companies are already stuck in a vicious cycle.

What do we mean by that? 

‍

Each time you use ChatGPT, you’re providing it with more data and boosting its ability to continue dominating the market. It’s even worse when you take into account that building solutions on ChatGPT is simply making it more difficult to compete with the model.

‍

It’s possible to escape this loop by exploring the power of crowdsource hardware in conjunction with community data. 

‍

Everyone can participate in this revolution, but only if we utilize existing hardware to make it happen and rely on an open-source community that doesn’t benefit just a handful of powerful companies. 

‍

No one wants to see a world where technology with gigantic potential to benefit humanity becomes a dystopian concept. The only way to stop the monopoly is for all the little guys to band together and steer clear of the “if you can’t fight them, join them” mentality.

‍