I distinctly remember the year 2006, when “the cloud” suddenly became a buzzword in our industry. Until then, we simply called it distributed computing. So when people started asking, “When are we moving to the cloud?” it felt quite strange. I thought, aren’t we just renaming distributed computing? We were already “in the cloud”; the name had just changed.

Fast forward to November 2022, and it feels like déjà vu with the launch of OpenAI’s ChatGPT and similar generative AI services. It’s a bit different this time around, though. The shift now is about making generative AI accessible to everyone, whereas it used to be a much more exclusive field. But the sense of a simple rebranding, of an old concept suddenly gaining new attention under a different name, that feels very familiar.

Certainly, the transformation of distributed computing to cloud computing and the development of artificial intelligence leading to models like ChatGPT have common threads:

1. Accessibility: Cloud computing has brought previously expensive and specialized computing resources to a wider audience. Similarly, AI models like ChatGPT are bringing AI technology into the hands of many people. Anyone can interact with GPT-based models online without needing to understand the complex underlying technologies.
2. Scalability: Cloud computing can be scaled to accommodate the needs of different users, from individual consumers to large corporations. Similarly, GPT models can be scaled in terms of the size of the model (the number of parameters), and can be fine-tuned or adapted to a wide range of specific tasks or domains.
3. Economies of Scale: Cloud computing leverages economies of scale, lowering costs as more users join the cloud. AI development, particularly with models like ChatGPT, also benefits from economies of scale. Training a model is a significant undertaking, but once the model is trained, it can be used to serve millions of users.
4. As-a-service Model: Just as cloud computing has given rise to the concept of “as-a-service” (such as Software as a Service, Platform as a Service, etc.), AI models like ChatGPT also follow this pattern. They essentially provide “AI as a Service”, enabling developers to embed sophisticated AI functionality into their applications without having to develop that AI capability from scratch.
5. Data-centric: Both cloud computing and AI models like ChatGPT are driven by data. Cloud computing allows for storage, management, and processing of big data. Similarly, GPT models are trained on large datasets and their performance continues to improve as they are exposed to more and more data.
6. Innovation and Disruption: Cloud computing has revolutionized many industries by providing cost-effective and scalable solutions. AI models like ChatGPT are similarly causing disruption, by automating tasks that previously required human intelligence and enabling new types of applications and services.

While there are significant parallels, it’s also worth noting some differences, particularly around interpretability and ethical considerations. These are areas that are very important in AI and are the subject of ongoing research and debate. Follow me on LinkedIn or Twitter