These three - Deep Learning, Personalization and Privacy are indeed oxymorons. The fact that when they come together for contextualization, relevance and differentiated customer experience - privacy takes a back seat! When Privacy takes centre stage, then Deep Learning & personalization take a backseat! The real question, is from a customer point of view, can they work in harmony? How do companies use them effectively?
What deep learning can do is well documented. When you listen to this talk by Jeff Dean on Large scale deep learning at Google, you will realize the power of its capability. The most important difference you will see is how Google has embedded that in its products and services.This then brings the issue of privacy to the center of the debate, as this is the data of the customer that Google analyzes & monetizes. However, when it comes to the synergy between Deep learning and personalization, all the data that is used to understand this customer, effectively means personalization can be done like never before. So, the next time you walk-in to your favorite store, or your bank, or your online store, or your health care provider - the personalization possibilities thro' virtual assistants can create a phenomenal experience to the customer. Hence, if deep learning helps customers make their life simpler, then what's the problem with privacy? I think if this is done with the permission of the customer, then puts to rest the debate.
However, this has a lot of learning for the other industries - as the customer pays and owns their products, unlike Google where products are sometimes used for free. Take for example an automotive brand - the applications of deep learning, personalization is phenomenal. With most cars fitted now with navigation systems and telematics devices, the data this can throw about the customer, their driving habits into their automotive platform is phenomenal. The driving habits, the engine data, travel data, innovative use of public data all represent great deep learning application opportunities for the automotive brands to drive personalization & differentiation. Here, privacy is something of little consequence or debate as any assisted customer service will only enhance the customer experience. The same goes with banks - where the usage of the mobile app, online banking, payment apps, shopping data thro' credit cards, public financial data etc. will only create opportunities for deep learning applications and personalization for enabling personalized financial solutions.
The question really is not will they but they will have to do it soon.