Internal surveys are pretty unreliable. Story time (details vague for obvious reasons):
I worked at a company where we were collecting safety compliance data on workers. The data was partially medical in nature, so we were legally bound by HIPAA, meaning we could not share the data with their employers. Some of the data we collected was self-reported, while some of the data was measured objectively via sensors.
Consistently, the self-reported data showed the workers were in compliance with safety standards. Consistently, the sensor data showed them to not be in compliance.
We changed the messaging around the self-reporting, to make it clear that a) we would not share self-reported data with their employers, and b) we were bound by HIPAA, and therefore could not share self-reported data with their employers. The number of workers self-reporting non-compliance increased significantly (more than doubled) but remained lower than 30%, while the sensor data continued to show compliance in single-digit percentages. The only thing we had achieved was objective proof that the self-reported data was unreliable.
The purpose of the self-reported data was to identify faulty sensors, and it was clear we weren't achieving that goal, so we eventually dropped the self-reported data from collection entirely.
Ultimately, there's no personal benefit to expressing discontent in an internal survey. Doing so takes the risk that the data will be used against you.
I do agree, there are some reasons to believe that people DO find satisfaction at FAANG job. It's just that there are reasons to believe the opposite as well, and I don't feel like I can draw a conclusion as confidently as the poster upthread who said, "They work in a field famous for people deriving immense satisfaction from their job". The only thing I'm convinced of at this point is, "I don't know."