While some call artificial intelligence in the form of ChatGPT ‘horribly bad‘ and don’t fear for their jobs or the future of humanity, others are both excited and scared by the possibilities. What will happen to artists, writers, and my own job if these AIs can be creative and challenge what we had considered deeply human capabilities?
Whatever the opinions and feelings about it, ChatGPT as the most prominent representative of the generative AIs that currently seem to overwhelm us with ever new applications, the fascination with it is undeniable. For example, OpenAI’s ChatGPT already had one million users in just 5 days of public deployment. In comparison, Instagram only reached this figure after 75 days. And by mid-May, just under half a year after release, there were 100 million users. What’s helpful is how easy all these AIs are to use. A browser is enough in most cases to use these tools. A similar boom occurred almost 26 years ago, when Netscape was released, the first browser that made the Internet accessible to many people for the first time. We know all too well how the Internet has fared since then. It is impossible to imagine our lives without it. It certainly changed our lives.
This also distinguishes AI from other technologies that were touted as the “next big thing that will change the world” and then disappeared as quickly as they came or are still stuck in the eternal cycle of “available in the next 10 to 20 years”. Flying cars, metaverse, crypto or eternal life are among them.
The framework drawn up by Venezuelan techno-economist Carlota Perez, in which she sets out three criteria that must be met in order to define a true technological revolution, proves to be helpful in this respect:
- Emergence of general purpose technology (multipurpose technology);
- Cheap key resources change the cost structure;
- Emergence of one or more infrastructures;
For example, in the 19th century, such a multipurpose resource was the steam engine, in the 20th century electricity. These made it possible to use them in all kinds of industries, such as mining, as motive power for machines in factories, or to drive locomotives. Low-cost coal, oil, or hydroelectric power became key resources that could dramatically lower the price of goods mined or produced and mobility, and increase manufacturing or transportation speed. At the same time, the necessary infrastructure, such as rail lines or power lines, was created, which made it possible to use it on a large scale and in all industries and societies for economic, military or private purposes. Computers in all possible forms are also such a multipurpose resource, whose key resources such as electricity, data storage or processors are becoming ever cheaper and more powerful and, thanks to the Internet and mobile communications, offer almost unlimited possibilities for use.
These technological revolutions are subject to a cycle already discovered by the Soviet economist Nikolai Kondratjev in 1926 and extended by the Austrian national economist Joseph Schumpeter. As each new multi-purpose technology is discovered or developed, the rate of innovation increases, then decreases until saturation occurs. As soon as a new multipurpose technology is introduced, the cycle repeats. The cycles benefit from the previous ones and the cycle durations decrease. Innovation happens more quickly, is implemented faster, and then falls off more steeply than in previous cycles.
And yet there is a big difference from the technology revolutions of the past. Their infrastructure took some time to emerge. Laying railroad tracks, building the power grid with distribution systems and generators, providing broadband lines, satellites and Internet connections took years and decades. Artificial intelligence as we are experiencing it here and as it has been dominating public discourse in many places since the end of 2022 is relatively ‘easy’ to set up and link to existing infrastructure.
This is exactly what gives us a Cambrian explosion of AI applications, which we previously only knew from smartphones and their apps and is now surpassed many times over.
Announcement: New Book
My second book on artificial intelligence will be published at the end of November 2023. After When Monkeys Teach Monkeys, which was published in early 2020 and in which I already described autonomous agents as mentioned here, Creative Intelligence: How ChatGPT and Co Will Change the World is coming to bookstores. In it, I will talk about generative and autonomous AIs, how they came to be, how they work, what they can and cannot do, how they are being used, and how they are changing our lives.