That's a misreading of Chenowith's argument which itself is heavily based on Timur Kuran's Revolutionary Thresholds concept.

The thesis is once mass mobilization of non-violent protesters occurs, it reduces the threshold for elite defection because there are multiple different veto groups within a selectorate, and some may choose to defect because they either view the incumbent as unstable or they disagree with the incumbent's policies.

I also recommend reading Chennowith's discussion paper clearing up the "3.5%" argument [0]. A lot of mass reporting was just sloppy.

Tl;Dr - "The 3.5% figure is a descriptive statistic based on a sample of historical movements. It is not necessarily a prescriptive one, and no one can see the future. Trying to achieve the threshold without building a broader public constituency does not guarantee success in the future"

[0] - https://www.hks.harvard.edu/sites/default/files/2024-05/Eric...