It's an interesting question to ask how likely it is to recover from civilizational collapse, and talking about 'stepping down in complexity' might be useful. I've previously only seen it discussed as whether we lose agriculture, science, or industry (see e.g. Baum et al., 2018). It seems the author is implicitly referring to the Energy-Complexity Spiral by Joseph Tainter, a fascinating concept:

The common view of history assumes that complexity and resource consumption have emerged through innovation facilitated by surplus energy. This view leads to the supposition that complexity and consumption are voluntary, and that we can therefore achieve a sustainable future through conservation. Such an assumption is substantially incorrect. History suggests that complexity most commonly increases to solve problems, and compels increase in resource use. This process is illustrated by the history of the Roman Empire and its collapse. Problems are inevitable, requiring increasing complexity, and conservation is therefore insufficient to produce sustainability.

It seems most x-risk scholars believe the probability of recovery is really high (>90%) as long as something like the scientific method is preserved (last few people problem). I think this is likely to be correct, and that the failed recoveries are either by extinction (70% of failed recoveries) or by loss of the scientific method (30% of failed recoveries). Permanent loss of technology seems unlikely to me, as technological development offers many advantages and is observed in most cultures.