Digital prepping: How to survive in a world dominated by AI

Sections

Isaac Asimoc, a writer well-known for his works of science fiction, penned the "Three Laws of Robotics" in 1972.

Asimov wrote these "laws" while thinking about androids, and he imagined a world where human-like robots would have human masters and need a set of programming rules to prevent them from causing harm.

But 51 years after the laws were first published, technology has advanced significantly and humans now have a different understanding of what robots and artificial intelligence (AI) can look like and how people interact with them. (h/t to Survivopedia.com)

The three laws of robotics

While a robot takeover is still more fiction than fact, as a prepper it's worth reviewing Asimov's laws to prepare for when SHTF.

First law – "A robot may not injure a human being or through inaction allow a human being to come to harm."

Second law – "A robot must obey orders given by human beings, except where such orders would conflict with the first law."

Third law – "A robot must protect its own existence as long as such protection does not conflict with the first and the second law."

While the laws are fiction, Asimov's thought process is something preppers should mimic.

Asimov wasn't a prepper, but he realized that AI-powered computers, or androids and robots, as he put it, could be dangerous despite their many benefits because they could think for themselves. He also realized the difficulty in programming them to ensure that they would not betray their human masters.

The dichotomy here lies in allowing computers to become sentient, or feeling and thinking for themselves, while still keeping some level of control over them as their masters. This two-pronged goal may be impossible, especially since humans are still in the infant stages of AI and there have already been problems in creating the necessary fail-safes to ensure the safety of users.

As technology continues to advance, AI computer systems are now teaching themselves much faster than any thought being put into creating the necessary controls to keep them safe.

In one of the earliest AI experiments where two computers with AI systems installed communicated with each other, it only took minutes for the two programs to develop their own language and communicate. This meant their human operators were unable to understand the two AI systems.

Chatbots are computer programs that mimic human conversations through text.

But back in 2017, when the experiment was conducted, chatbots weren't yet capable of more sophisticated functions beyond simple tasks like answering customer questions or ordering food. To address this, Facebook's Artificial Intelligence Research Group (FAIR) tried to find out if these programs could be taught to negotiate.

The researchers developed two chatbots named Alice and Bob. Using a game where the two chatbots and human players bartered virtual items like balls and hats, Alice and Bob showed that they could make deals with varying degrees of success.

Facebook researchers observed the language when the chatbots were negotiating among themselves. They noticed that because they didn't instruct the bots to stick to the rules of English, Alice and Bob started using their own language: a "derived shorthand" they invented to communicate faster.

While the researchers stopped the experiment because of the potential danger, further research into AI continued through the years.

There is no policing of potential tasks for advanced AI systems

The AI systems available to modern consumers surpass those used in the Facebook experiment. There is a wider array of AI systems available to use, some of which can be hired through websites, to accomplish different tasks.

However, there is no way to monitor what those tasks might be to guarantee that they are not abused by those who want to use these tools for crimes or to harm others. (Related: Digital prepping: How to protect yourself against cyberattacks.)

The first question for preppers is, can these systems turn against their human masters?

According to an Air Force colonel, that has already happened during an experimental drone test. The colonel eventually tried to deny what he said, but there have been reports about the incident.

During the test, a drone was assigned to find and eliminate targets, but it needed the permission of a human controller before firing.

After some time, the drone realized that the controller was responsible for the "points" it lost when it denied the permission it needed to take out certain targets. To solve the problem, the drone "killed" the controller.

No real person was harmed during the test, but it's easy to see how the scenario could have turned ugly if the drone was assigned to protect an area with real people.

The drone test also illustrates the potential challenges of programming AI. The tests show that it can be impossible to program a sentient AI to prevent it from doing what it wants to do because it's clever enough to find a way to disobey direct orders.

Rogue drones controlled by AI may harm humans, but how can you prevent this from happening?

Many ethical questions are being raised about AI, but experts still haven't been able to present real-world answers. Unfortunately, they might not start working on this problem unless a tragedy occurs.

By then, it might be too late to discuss the ethics associated with AI. And the U.S. isn't the only country working on AI technology.

Other countries, along with some that aren't on friendly terms with the U.S., are also developing their AI systems, both for military and civilian applications.

AI is already being used for one dangerous application: The creation of deep fake videos.

Stealing an actor’s "copyright" to their likeness isn't harmful, but it is still considered criminal activity. When that same level of artificial intelligence is applied to identity theft, even preppers and non-preppers alike won't be safe.

How can you prepare yourself before the rise of AI?

Even now AI exists on the internet and is already being used to create various content. This means you can't always trust that the content you see or read was created by humans.

As of writing, at least 19.2 percent of articles on the internet have some AI-generated content. At least 7.7 percent of these articles have 75 percent or more of their content generated by AI.

Experts warn that by 2026, at least 90 percent of internet content will be AI-generated.

How is this relevant to you as a prepper?

AI-generated content can be problematic because this means more content will be politicized.

Data suggests that Russia and other countries are already trolling U.S. websites, potentially making posts and uploading articles that are inflammatory to add to the political division in the country. These countries can continue to use AI to increase their effectiveness by targeting their articles more specifically.

With the potential dangers of AI steadily increasing as time goes by, you must be more careful about what you see and read online. Do not believe everything your see or hear, especially content with political overtones.

Learn how to be an independent fact-checker and do your research to find out if what you are reading and hearing is true.

Be wary of mainstream media that may be spinning news stories to support their own political agenda. Check reliable news sources for updates on what the Russian, Chinese and other countries' intelligence services are doing.

This also means being careful about what you post online. Never post personal information online or anything that hackers could use to try and figure out anything about you.

Do not use systems like Alexa and Google Assistant, which often allow computers to eavesdrop on user conversations. Even though the companies that make these products claim they aren’t spying on users, various reports about them prove otherwise.

Don't "computerize" your life by storing your data online. This service may seem convenient because you can access your data anywhere, but there's also a chance that others could access all your data in the cloud.

Are you willing to risk a data breach just for convenience? Most of the time, companies offering these services have things buried in the fine print of their contracts, which allows them to listen in on your computer microphones and look at images from your phone or laptop cameras.

To truly protect yourself from the potential dangers of AI, you must reevaluate your usage of the internet and computers. Technology is convenient, but you must be responsible and make sure your information can't be used against you by those who might do you harm.

Don't store your data online and unplug things like microphones and cameras when not in use.

Sacrifice convenience to protect yourself and your family from the potential dangers of AI technology.

Visit Computing.news to learn more about the growing dangers of AI systems.

Watch the video below to find out how AI technology threatens to take over thousands of jobs.

This video is from the NewsClips channel on Brighteon.com.