In a double-blind, phase 3 trial only one group has on-going data access: the independent group conducting the study. No one else, including patients, doctors, and the business, has any knowledge of the results until they are "unsealed". That continues to be true throughout the entire trial even if preliminary results are unsealed at different points.
Precisely. They have seen 94 infections total so far, based on the statistics that’s likely to be almost entirely in the control group. Those infections are reported as they come in, so for example, perhaps last week there were 87 infections. The study ends once they hit ~150 infections, based on their sample size that will give them the confidence intervals they need to finalize.
So the difference between announcing preliminary results last week versus this week was their own internal decision on timing.
