This Man Taught Google Assistant to Shoot a Gun and the A.I Learnt it Quite Well

0
525

The dreaded future most of us have thought about A.I is almost coming to reality. This is the future where everything we’ve seen intelligent machines do in Sci-fi movies will start taking place and humanity will be enslaved by robots.

Meanwhile, at the forefront of this race to a dead end reality is Google assistant.

With Google adding so many features to the digital assistant; puts it further to become the type of A.I we’ve been seriously nursing fears over.

Google assistant can now do so many things, from normal tasks like turning on your lights and setting reminders for you to convincingly mimicking human speech patterns.

The things that this AI helper is so capable of doing this days is just so scary.

Its latest (unofficial) ability, though, is a bit more sinister than you can possibly imagine. That’s the Google duplex thingy.

Recently, an Artist by name Alexander Reben taught Assistant how to fire a gun and Google assistant learned it quite well.

While in progress, the victim of this experiment was an apple, (not Tim cook’s Apple) but the fruit apple and for sure, it wasn’t a living being neither.

In a 30-second video clip, simply titled “Google Shoots,” Reben could be seen commanding the Assistant, “OK Google, activate gun.”

Then, barely a second later, a buzzer goes off, the gun fires, and Assistant responds “Sure, turning on the gun.”

On the surface, the footage is underwhelming and will sure raise Alot of controversy as to where is the boundaries being set for A.I could currently be located?

Reben stated, “the discourse around such a(n) apparatus is more important than its physical presence.”

He said he just choses to use Google Assistant for this project, but it could also be replicated on Amazon Echo “or some other input device as well.”

As seen in the video, Reben was the one who commanded Assistant to “activate gun.” Which means is still the person responsible for the action.

But the project will strike up controversy in the A.I field when people start considering the fact that using machine learning these days, A.Is are smart enough to anticipate our needs and cater to our comfort every day, it’s not hard to imagine a day when digital assistants could kill off people who upset us, if given access to weapons. Almost looking like the Sci-Fi movies that we’ve all feared A.I technology will evolve into.

A.I have found its way into so many things in our lives and they can only continue spreading no going back again. We have A.I in our cars, smartphones, industrial machines and so on and even in the military.

We can demand companies to make sure that their tech cannot cause harm, but it’s simply impossible to plan for every eventuality, there are every possible reason to think and predict that AI might go rogue someday.

Google might not ever have imagined Assistant being used to shoot a weapon. But it took Reben just a couple of things to achieve that. According to him, he used parts laying around his studio.

He used a control relay that usually turns on a lamp, linked it to a Google Home speaker, then connected a laundromat change-dispensing solenoid to pull a string that looped around the trigger. “The setup was very easy,” Reben said.

We already know that AI hears, sees and can act but Reben has shown us that it can be tweaked further to harm people and bring the doom we’ve all feared closer to reality.

Imagine what could happen if nefarious programmers and the vilian-like individuals in the movies pick this idea up, tweak it into an AI that has gone rogue. I think this is the time we all need to have this important discussions about how, and if we can possibly prevent an intelligent machine from killing us all and endangering the existence of the human race

LEAVE A REPLY

Please enter your comment!
Please enter your name here