That makes me think, if they can't tell we're humans or bots while they're harvesting our data, how easily could we poison their data with fake people generating false data?