Cambridge, MA – I’m fresh out of an academic keynote panel here at the MIT CIO Symposium titled, What Every CIO Should Know About the Future Impact of Digital Business.
From what I took away, a big chunk of the CIO horizon revolves or is about to revolve around customer or employee privacy, data security, risk and regulation.
That’s not so surprising, given the mix of new internal and startup business ideas presently running headlong into information glut. As someone pointed out, the command and control business model is headed for extinction. To the degree that's true we’d better figure out a new framework that increasingly might be what CIOs come to manage and monitor on the technology side – with business guidance.
David Clark, Senior Research Scientist at MIT CSAIL pointed out that the telephone, music and video industries have already had their “violent collision” with the Internet and that risk now spreads all the way from architecture to the consumer.
Most revolutions have been preceded by evolutions in measurement, said Erik Brynjolfsson, Director of the MIT Center for Digital Business. This is now catching up with the use of information where “the revolution affecting managers and CIOs today is the ability to measure data in a fine-grained way.”
Sandy Pentland, who heads up MIT’s data measuring Media Lab, put a more blunt point on it. “If I have non-public data sources, there are things I can do that nobody else can do.”
If personal and private data is becoming the new cow path to competitive differentiation, Pentland figures the CIO workday of tomorrow is going to veer between increasing volumes of big (personal) data – and keeping customers and regulators happy.
Nor do I. According to the panel, the new framework will arise from giving customers and employees more control of their personal data and how they incrementally release it in a quid pro quo kind of way.
Any number of outlets (information-management.com included), do this sort of thing every day in return for extra value content, but in a broader sense, Pentland thinks granular control of standard profile information might offset some unintended consequences.
The example was given of European Union instituted privacy regulations that made it very restrictive for companies to track behavior for custom advertising. Sometime later, Brynjolfsson said, ad effectiveness declined significantly and media companies needed more revenue. More ads, interstitials, pop-ups and every other kind of annoyance came to make consumers worse off rather than better.
Infrastructure to execute granular permissions doesn’t widely exist yet, but testing shows it will scale, Pentland says. The data Armageddon arrives, he says, when the infrastructure arrives and all companies start doing the same thing (using it).
That will make companies more dependent on their analytic skills, and probably lead to some major data disasters with unintended consequences like we’ve seen in financial markets.
As for liability, Pentland even suggested offshore banking as a kind of allegory for cloud computing where the data doesn’t really exist anywhere, though Clark said people once thought the Internet had no extrajurisdictional boundaries. (I thought of a glaring jurisdiction in China and more recently, the fate of online poker sites.)
“We’ll come to rely on algorithms and they’ll interact and people will become vulnerable,” Brynjolffson said, offering the harmless example of a $2.5 million book for sale on Amazon that came to exist as a result of competing price algorithms run amok.
In the end, it’s a mix of data granularity and good analytic skills, the panel seemed to be saying. We’re not sure where it ends but will be back next year to find out.