5 ways AI can help solve the privacy dilemma

There is no disputing the privacy trend. It is here. It is unstoppable. And it is one of the few issues in American life that crosses party lines.

Data shows that 86% of people care about privacy for themselves and others — with 79% willing to act on it by spending time and money to protect their data. And to those cynics who say people moan about privacy and do nothing, the same study found that 47% have taken action because of a company’s data policies.  

What does this mean for the trillions of dollars that flow through the U.S. economy as a result of the very same privacy violations that are enraging consumers? It appears to be a tectonic conundrum; consider that Meta conceded that Apple’s change in its privacy rules has and will cost them billions.

But for companies suffering from the effects of Apple’s shift to opt-in from opt-out, artificial intelligence could be a solution.

Protecting privacy while allowing the economy to flourish is a data challenge. AI, machine learning, and neural networks have already transformed our lives, from robots to self-driving cars to drug development to a generation of smart assistants that will never double book you.

There is no doubt that AI can power solutions and platforms that protect privacy while giving people the digital experiences they want and allowing businesses to profit.

What are those experiences? It’s simple and intuitive to every Internet user. We want to be recognized only when it makes our lives easier. That means recognizing me so I don’t have to go through the painful process of re-entering my data. It means giving me information — and yes, serving me an ad — that is timely, relevant, and aligns with my needs.

The opportunities within the “personalization economy,” as I call it, are vast. McKinsey published two white papers about the size of the opportunity and how to do it right. Interestingly — and tellingly — the word “privacy” isn’t mentioned a single time in either of those white papers. That oversight is remarkable and overlooks the tension between privacy and personalization.  If you accomplish the former but sacrifice the latter, millions of consumers will miss out.

AI can find that balance — a privacy-respectful personalization, or PRP, and can satisfy our hunger for personalization and recognition, which is hard-wired into the human brain.  

Here are five ways AI can achieve a new era of PRP to enable relevance and support entrepreneurs and brands at the same time.

1. Create platforms that can process millions of inputs in real-time — contextually, behaviorally — with no user ID and no cookies. 

Importantly, this must be done in a way that requires no coding or friction and can easily be deployed on the open web. 

In this way, AI will make sure that the unstoppable privacy trend doesn’t disenfranchise those who need to find an audience for their product or service. 

As part of this AI revolution, there is an opportunity to create a “privacy seal” — one that goes beyond the must-have limits of compliance with GDRP and other requirements (like TrustArc) and lets the 79% of consumers who will drop their engagement with private-invading brands know that an ad unit is free of cookies.  

2.  Use AI to generate first-party data.

We know that people will share data if there is true reciprocity in that relationship. For example, an EV manufacturer can use a PRP platform to advertise its cars and bring people into the marketing funnel, but the ultimate goal is to find the most qualified and interested EV buyers at the top of the funnel.  

A sophisticated and trained AI model can convert leads that are generated through PRP advertising first-party data by processing the right signals — daypart, geography, time spent on ad, for example — as well as asking the right questions in a quiz to discern intent.

And rather than have copywriters scribbling away to create thousands of testable messages, marketers can use GPT-3, an AI-led deep-learning model that produces human-like text, to inspire the reciprocity.

3.  Use AI to motivate, engage and connect with first-party data. 

The same AI that can build first-party databases can and must be leveraged to connect with users in a more targeted manner.  

However, this is not currently taking place. Database marketers still send full-file emails when the ability exists to use AI to truly recognize me as an individual. If the industry doesn’t use AI to create a new era of PRP, then marketers will continue to underleverage the databases they have spent billions to build.  

In other words, build an AI-driven flywheel that drives revenue in a privacy-first world.

4.  AI can enable marketers to better leverage influencers and creators

This is an exploding category that is transforming marketing and that, in part, enabled TikTok to generate more traffic than Facebook last year. Recently, Triller bought Julius, combining an influencer platform with software tools in an AI-led combination that will sharpen the ability of brands to find relevant influencers. This is a way to generate even more first-party data, all within the walls of PRP.

5.  Beyond pure marketing, AI plays a huge role in protecting privacy by securing data from hackers. 

One of the most exciting applications of AI is in the field of adversarial learning. Consumers are looking at privacy holistically — not just what marketers do by selling their data and tracking them, but also their efforts in securing their data from breaches.

Human nature has never changed, nor will it. We want to be recognized as unique individuals, and we have highly-tuned instincts when it comes to protecting ourselves and our families from unwelcome and uninvited intrusions. AI can make sure both of those needs are met, and marketers who get to PRP first will lead their businesses into the future.

Doron Gerstel is CEO of Perion.