This is called risk selection, and is already a common practice [1] that doesn't require much in the way of "health data" [2]. I suspect that 80% of the effect

This is called risk selection, and is already a common practice [1] that doesn't require much in the way of "health data" [2]. I suspect that 80% of the effect can be achieved via "simple" methods like advertising on health and fitness related sites or keywords, partnering with gyms to advertise to members, etc. ACA has some provisions to help address these incentives [3], though it's unclear how effective they are. The Privacy / Big Tech angle is attention grabbing, but is grossly missing the point. In my opinion, the reality is we need to decide if this type of discrimination is something we as a society want to regulate or not. [1] https://www.nber.org/papers/w19198 https://www.nber.org/papers/w19198 [2] https://www.federalreserve.gov/econresdata/feds/2015/files/2015101pap.pdf https://www.federalreserve.gov/econresdata/feds/2015/files/2... [3] https://www.kff.org/health-reform/issue-brief/explaining-health-care-reform-risk-adjustment-reinsurance-and-risk-corridors/ https://www.kff.org/health-reform/issue-brief/explaining-hea...