On 25.3.26, the FDPIC published on his web­site Notes on weara­bles published. They are aimed at users – buy­ers and par­ents who track their children.

The fol­lo­wing requi­re­ments and recom­men­da­ti­ons can be deri­ved from the information:

Manu­fac­tu­rer and supplier

  • Pri­va­cy by Design
  • Encryp­ting data transfers
  • Secu­ri­ty updates, vul­nerabi­li­ty assessment
  • Pur­po­se limi­ta­ti­on: Use of data for mar­ke­ting or pro­duct deve­lo­p­ment only with express consent
  • Trans­pa­rent data pro­tec­tion declarations
  • Desi­gna­te cont­act per­son for data pro­tec­tion inqui­ries; repre­sen­ta­ti­on in Switz­er­land if necessary

User (buy­er)

  • Read pri­va­cy poli­cy and terms and conditions
  • Check whe­re data is stored
  • Rest­rict app per­mis­si­ons to what is func­tion­al­ly neces­sa­ry, reject or revo­ke unneces­sa­ry ones
  • Install updates regularly

Users of smart glas­ses and came­ra-enab­led wearables

  • Informing third par­ties about recor­dings and obtai­ning consent
  • Refrain from under­co­ver recor­dings (cri­mi­nal law)

Par­ents

  • Act in the inte­rests of the child and respect their privacy
  • Child­ren can­not con­sent to their own surveillance

You will lar­ge­ly agree with this, except on one point:

Under Swiss law, the use of data for mar­ke­ting pur­po­ses or for the deve­lo­p­ment of own pro­ducts requi­res the express con­sent of the data sub­ject (see also Coo­kie Guide).

The pro­ce­s­sing of par­ti­cu­lar­ly sen­si­ti­ve per­so­nal data only requi­res con­sent if the prin­ci­ples of data pro­ce­s­sing are vio­la­ted or such data is pas­sed on to other data con­trol­lers (see e.g. here). The con­trol­ler may the­r­e­fo­re gene­ral­ly use health data for mar­ke­ting pur­po­ses and pro­duct deve­lo­p­ment wit­hout con­sent, inclu­ding with pro­fil­ing, also using machi­ne lear­ning.

The FDPIC does not justi­fy the con­sent requi­re­ment in his gui­dance, but is likely to con­sider the pro­ce­s­sing of health data for the pur­po­ses men­tio­ned as dis­pro­por­tio­na­te from which he deri­ves a requi­re­ment for consent.

Whe­ther this argu­ment is admis­si­ble or rather the free pur­po­se of the per­son respon­si­ble vio­la­ted, does not need to be dis­cus­sed in detail here (see here). In any case, howe­ver, dis­pro­por­tio­na­li­ty could be only in indi­vi­du­al cases and all cir­cum­stances would have to be taken into account, including

  • the cost of the weara­ble or the asso­cia­ted services,
  • the que­sti­on of whe­ther or not sen­sor data for mar­ke­ting pur­po­ses uses poten­ti­al health data as such. Upsel­ling in an app based on trai­ning data, for exam­p­le, uses data that allo­ws con­clu­si­ons to be drawn about the sta­te of health, but the con­trol­ler does not have to use this infor­ma­ti­on con­tent. If he does not do so, his pro­ce­s­sing can­not be equa­ted with the pro­ce­s­sing of health data;
  • Opt-out or other con­trol opti­ons for the user;
  • whe­ther pro­duct deve­lo­p­ment or mar­ke­ting mea­su­res are also in the well-under­s­tood inte­rest of the user.

It is also inte­re­st­ing to note the FDPIC’s refe­rence to the Coo­kie gui­de. The FDPIC assu­mes that the use of non-essen­ti­al coo­kies tends to be dis­pro­por­tio­na­te and then requi­res justi­fi­ca­ti­on. This is que­stionable in such gene­ral terms. In any case, howe­ver, the FDPIC also express­ly lea­ves out the Justi­fi­ca­ti­on by over­ri­ding inte­rests open, and the same should app­ly here.

As a result, tho­se respon­si­ble for data pro­ce­s­sing via weara­bles are advi­sed to at least pro­vi­de an opt-out right, i.e. a low-thres­hold opti­on for limi­ting the use of sen­sor data. In this case, over­ri­ding inte­rests are more likely or – which amounts to the same thing – a clas­si­fi­ca­ti­on as proportionate.