Generally speaking, research is partitioned into B2C ('general population') and B2B (e.g. 'VP of Finance at a Fortune 500 company'). Around twenty years ago, API-based vendors started supplying participants, and this coincided with survey methodology moving from in-person to online.
High incentives for B2B studies invites a common source of fraud today: survey farms programmatically using LLMs to get sufficient context on quota criteria and then piping responses into the online survey. It is not abnormal to see a 40% fraud rate in a B2B survey. B2C is problematic, too, but less so due to lower incentives.