Although it might puzzle or even infuriate data scientists, suspicion about big data is understandable. The concept doesn’t seem promising to most people. It seems scary. This is partly because big data research is shrouded in mystery. People are unsure about organizations’ motives and methods. What do companies think they know about us? Are they keeping their insights safe from hackers? Are they selling their insights to unscrupulous parties? Most importantly, do organizations use our personal information against us? Big data research will only overcome its suspicious reputation when people can trust it.

 Some scholars and commentators have proposed review processes as an answer to big data’s credibility problem. It is possible that a review process for big data research could provide the oversight to ensure the ethical use of data we’ve been hoping for, applying sensible procedural rules to regularize data science. But procedure alone isn’t enough. In this essay, we argue that to truly protect data subjects, organizations must embrace the notion of trust when they use data about or to affect their human users, employees, or customers. Promoting meaningful trust will involves structuring procedures around affirmative, substantive obligations designed to ensure organizations act as proper stewards of the data with which they are entrusted. To overcome the failures of a compliance mentality, companies must vow to be Protective, Discreet, Honest, and above all, Loyal to data subjects. Such commitments backed up by laws will help ensure that companies are as vulnerable to us as we are to them. When we know we can trust those using big data, the concept might not seem so scary after all.  We will disclose more and more accurate information in safe, sustainable ways. And we will all be better off.

Download