AI technologies, including generative models such as ChatGPT, classification algorithms, and optimization algorithms, serve as a powerful complement to Augmented Collaborative Learning (CAA), but they can neither replace nor should replace it.

Their strength lies in their ability to incorporate virtual experts and utilize large amounts of data (big data) to enrich collaboration streams, thus supporting collective intelligence. They can also be used to detect and even generate useful patterns and extract relevant information from complex collaboration flows.

However, these technologies cannot take over human collaboration in the context of CAA. The collaboration between human beings, each with their complexities, mindsets, and unique experiences, cannot be replicated by machines in most cases, and should not be in many cases. Indeed, responsibility requires free will; AI programs can at best emulate this freedom, but they never have the "free will" necessary to make a decision and assume its consequences, nor to fully commit.

Moreover, as AI approaches "general human intelligence," it will have to integrate corresponding values and beliefs, filter perceptions based on these values, set aligned goals, and make decisions consistent with those goals. This means that different communities, such as countries, organizations, or groups of scientists, will want to create their own AI systems. Since there is not one "general human intelligence" but rather a multitude of diverse human intelligences, there will result in a variety of artificial intelligences that must collaborate with each other and with humans.

In this sense, CAA, as implemented by SymPlace, will take on greater importance and presence, as it will serve as a foundation for cooperation between humans and algorithms when it comes to making crucial decisions.