Here’s How Siri Could Become Your AI Cyber Nanny

3 minute read
| Editorial

The internet has turned into the Wild, Wild West. People are exposed to threats daily, but help is often far away in time and space. But, like the old American Wild, Wild West times and technology change. It’s high time our leading tech companies like Apple and Microsoft put artificial intelligence to work truly protecting us. That’s the noblest cause for advanced technology right now.

_____________________________

The recent security threats against macOS (handbrake) and Windows XP (ransomware) reminded me of something intriguing. Namely, we spend a lot of time thinking about how artificial intelligence (AI) agents might serve us when to comes to selling us things, playing music or providing answers to information queries. However, there’s very little discussion about how AIs can defend us from the Wild, Wild Internet.

Gunfighters

Hey kids. Come play with us.

The Real West

There was a time in the American West, in the 1800s, when adult males carried a holstered firearm. In most situations, the gun was used for defense against wolves or bears. Or perhaps it might be needed to humanely put down a seriously injured horse. Or defend against robbers. Or worse.

In more extreme cases, grown men with a talent for handguns drew on each other, often for petty reasons. Sometimes for important reasons. But, by and by, as a society we decided that local law enforcement was a better idea, and in 50 years or so it became generally unnecessary for grown men to always carry a gun. The violence had to stop.

Today’s internet is like a small western town in the mid-late 1800s when drunken outlaws would ride into town, shooting up the place, and the streets were no place for children.

Today, any child or adult with a computer, tablet or a smartphone is a target. Sensible frontier mothers would keep their children off the streets and away from the saloon on Saturday nights, but in 2017 we happily hand our children an iPhone and send them on their way, hoping for the best. But grown men draw on them every day.

Proper Defenses

Now Apple does a pretty good job of defending us, but it seems, most times, like an endless game of whack-a-mole. The average adult is too busy living, working and playing to become an internet security expert. Lapses happen.

I particularly blame Microsoft for not having a Windows XP kill switch. Allowing this ancient OS to survive for the last 15 years is just not a prudent, ethical practice. I know. The principle, it can be argued, is that some people don’t like to have their hands forced. But it’s like telling a customer that their airbag is defective and could kill them. A fix is socially mandatory or people get hurt.

We need a new kind of help. Siri needs to chart a new, more imaginative path.

That Stupid OS

There are what I call dumb defenses. Apple’s OSes have a system called XProtect, ASLR, sandboxing and other techniques. Users can also add malware protection, if they know which one to use and how to use it effectively. However, by and large, the protection systems on an Apple (or any other) computer are procedural. They work, if well used, in the simplest cases, and bring to bear no real interactive intelligence.

I think it would be great if Apple used its enormous resources and talent with AI to bring a new Siri sheriff to town. Smart defenses. If an AI is going to pry into our lives and preferences, it mights as well go the whole way and start to develop a genuine, symbiotic relationship with our us and our OSes.

Modern OSes have tens of millions of lines of code. Their operation is opaque to us. We tend to cross our fingers and tediously patch. If we have time. We apply only the crudest tools to protect us from daily assaults, and when infected, we need serious counsel.

An AI that can really dig in, talk to us, mentor us, scan our systems, advise against phishing assaults, and guide us away from dangerous websites and actions may be the only way to escape from the Wild, Wild Internet and its endless outlaws on the streets, making life miserable for everyone.

A Grand Challenge

There are the grandest challenges for our species. Things like protecting this planet, the only home we have. Colonizing Mars. Curing disease. Harnessing nuclear fusion and launching an era of plentiful, cheap energy. Securing the national infrastructure.

And yet. When it comes a grand challenge to block the endless exposure to daily threats from the Internet, Apple (and others) remains generally silent and settles for an endless, almost futile series of OS patches that barely hold the fort.

Whack-a-mole security.

Forget the Knowledge Navigator. I want an intelligent, communicative, security navigator who’s a genuine AI partner in this endless battle. It’s a grand challenge that Apple is perfectly suited for.

______________________

Gun flight at OK Corral via Shutterstock

3 Comments Add a comment

  1. wab95

    John:

    Well-said, and timely. A quick observation and recommendation.

    A chief barrier to this is less likely to be technological than it is to be inertial. Successful solutions to grand challenges are about paradigm shifts. This requires that, first, we recognise the challenge. Second, we summon the imagination to see a plausible solution that is unorthodox, novel, even counter-intuitive for which the complete package of technology might not even exist, and third, to have the courage and the drive to pursue it in the face of inevitable resistance.

    Studies have shown that nearly 75% of research is mere repetition of what has already been done, just incremental advances on a theme. Only 25% is original, and the barriers to achieving it are more than just a failure of imagination, but the activation energy required to overcome disciplinary or environmental inertia that sustains the status quo and resists change.

    A part of that inertial environment, for a company that sells product, like Apple, is consumer expectation and preference. If, as has been the case with many of Apple’s attempts to harden OS security, whether by hiding sensitive folders from inexpert and inadvertent consumer tampering, let alone tampering by the bad guys, or not permitting app side loading, or any number of relatively small incremental changes, many consumers complain about the Nanny State of Apple, what might consumer response be to a paradigm shift in which substantial security control is ceded to AI?

    Hazard and threat communication argues that, one of the greatest threats and anxiety provoking stimuli is to take control away from people, particularly something over which they once held principal dominion, particularly if that change is sudden and people have no choice but to accept it. Imagine getting into your car in the morning to go to work, and before you can press the start button, Siri’s voice comes over the comms system and says, ‘I have assumed control of this vehicle and I’ll be driving you to work today. Enjoy the ride’. Still feel like going to work?

    To effect such a paradigm shift as AI taking point on security, which if done with a high degree of efficacy, would be a good thing, Apple and MS and Google should do two things: 1) engage the community and communicate both the need and the transition plan for AI to assume greater control; 2) permit options during that transition period for consumers to wrest some of that control as needed, until – and this is critical – they come to trust AI with this new power.

    Once trust is built, then the paradigm shift is complete, and equanimity will be restored.

Add a Comment

Log in to comment (TMO, Twitter, Facebook) or Register for a TMO Account