Critics say that the cloud is really just a “an expensive computer in a another room”. Defenders argue that the cloud actually saves costs. I think discussion focuses too much on infrastructure as a service (storage, compute and networking).
It misses the fact that clouds like Microsoft Azure offer high-level services like Azure AI. Leveraging such technologies in-house would require a team of data scientists and engineers. Azure offers a service that does all the heavy lifting for you.
Here’s an example of one guy who built ChatCSV who built a basic web app on top of ChatGPT and sold his idea to a larger company. My message is that any manufacturer can implement technologies like computer vision or speech recognition to build a proof-of-concept without making a large commitment.
This is the biggest benefit of the cloud when it comes to data and AI: it allows non-IT companies to play with new IT-technologies as if they were LEGO-blocks to quickly decide whether AI is worth it or not. It abstracts away difficult mathematical concepts like model design similarly to how R or Python hide the difficulties of memory allocation and management in C.