To start with, I don't really know about AI, but I have worked in robotics, so I'll speak from that. There are cases where it's ambiguous to call something an emerging business. In fact, this technology is more than 20 years old, and existing companies have already staked out the market. So, apart from space, the industries you mentioned actually have quite a long history.(To be precise, it wasn't the robotics research field—I built my career specializing in installing robots and getting them running in factories)
What I mean by an emerging industry is one that requires little capital and has few entrenched players—in the end, an industry where new entrants can compete with small initial capital. That's how I'd redefine it.
Because in the IT boom era, there were many technological opportunities. The boss who hired me was one example. But in the case of industrial robot manufacturing, the existing hardware supply chain is, by and large, the one investing.
In the end, what I want to say comes down to this: there are fewer industries open to new entrants.
Because in the IT boom and the construction boom, there were markets that needed less capital than you'd think in the early days. In subcontracting, for instance, you could enter with relatively little capital. But in today's technologically advanced industries, initial capital requirements are far higher. So in the past, even within the same category of industry, there clearly existed sub-markets you could participate in with comparatively little initial capital.
For example, in the old IT industry, my seniors 20 years ago said it took surprisingly little capital. But today's market isn't like that. I think this is a problem that arises as industries become more advanced.
There's a paper I read some time ago arguing that as an industry becomes more advanced, the knowledge and resources required to make a product increase.[1]
[1]https://www.nber.org/papers/w27787