For the predictable reasons, the article overemphasizes "number of satellites" and under-emphasizes "height of satellites" and "inclination of satellites."
The CTC-1 constellation proposes to be at 510 km altitude and 97.4 degrees inclination[0], which is already a heavily-populated orbit[1] due to being in a Sun-synchronous orbit. Since the collision risk scales as the object density squared, this is an especially foolhardy decision from the perspective of space debris and space sustainability.
Remember that most of the satellite collisions occur in a "halo" around the North and South poles where the SSO orbits all pile up. Avoiding these orbital slots (and in fact, removing defunct objects from these valuable orbits) is the best thing we could do for Kessler syndrome. China is doing literally the exact opposite.
It also doesn't help that China just abandons their upper stages in orbit, rather than doing proper deorbit burns.[2] Since each Chinese rocket also can only launch a handful of satellites (vs almost 50 per SpaceX launch), the number of abandoned debris upper stages is truly massive, and again they're all being carelessly discarded in pretty much the worst possible orbit.
[0] https://forum.nasaspaceflight.com/index.php?action=dlattach;...
[1] https://forum.nasaspaceflight.com/index.php?topic=44021.0
[2] https://arstechnica.com/space/2025/10/everyone-but-china-has...
Lower and atmosphere drag becomes a problem to maintain orbit.
Higher adds communication latency.
With mega-constellation launches accelerating, the sci‑fi premise of imprisoning ourselves behind a debris field feels less fictional. This is essentially the collision-cascade risk described by Kessler Syndrome
Kurzgesagt has a good explainer. Hopefully we never trigger it.
As for the state of these networks, G60/Qianfan had a plan of ~650 sattelites by the end of 2025, but currently sits at 108. They hope for ~1200 by the end of '27
Just before the end of the year the GuoWang constellation hit 136 of their planned 13,000.
For reference starlink has launched over 10k satellites to date with ~9,400 in active service.
Im sure the constellations will grow, but they have been experiencing the pains of scaling, especially with 1 use rockets. SCMP loves to pump up these crazy plans and massive numbers as a national pride win, even when they are not feasible or still really far off.
If you add in EU providers, depending in how you count then, there's at least 2 or 3 providers who have more than 100 LEO satellites active.