If you look up the definition of intelligence, you get a statement about it involving the collection of information, knowledge, and data and applying critical thinking and logic, plus a lot of abstract consideration, to solve a problem. A seemingly imprecise result, but entirely a surprise, as the concept itself is complicated. If you look up the meaning of inference, you get a lot closer to what AI is really doing. Some define inference as steps in logical reasoning, moving from premises to logical consequences. There is in fact something called AI inference, which refers to the process where a trained machine learning model applies its learned knowledge to make predictions, decisions, or conclusions.

So this morning I thought I would engage in some mental aerobics. I was asking Bing’s Copilot AI a series of questions, which eventually led to the following question.

“I’m actually curious about reverse engineering what might have happened to Mars, working from the perspective that Mars was once Earth-like.

Furthermore, I imagine a civilization, not much unlike that of Earth’s current civilization, that was wiped out due to some type of catastrophe. Resulting in the loss of the outer layers of organic soil and water, leaving no hint of their existence. What type of event might cause such a result?”

This was the answer it provided.

**A combination of planetary-scale catastrophes—like magnetic field collapse, atmospheric stripping by solar wind, and massive impacts—could theoretically erase signs of a once Earth-like Mars, including any surface civilization.**

Let’s reverse-engineer Mars’ transformation from a potentially habitable world to the barren planet we see today:

—

### 🌍 Step 1: Mars as Earth-like

– **Ancient Mars had liquid water, a thicker atmosphere, and possibly a magnetic field**, all essential for habitability.
– Geological evidence suggests rivers, lakes, and even oceans existed during the Noachian period (~4 billion years ago).
– If life or civilization had emerged, it would have needed protection from radiation and climate stability—both provided by a magnetic field and dense atmosphere.

—

### 💥 Step 2: Catastrophic Collapse

Several interlocking events could have wiped out a surface civilization and erased its traces:

#### 1. **Magnetic Field Collapse**
– Mars’ core cooled early, halting its dynamo and magnetic field.
– Without this shield, **solar wind stripped away the atmosphere**, exposing the surface to intense radiation.

#### 2. **Carbonate Catastrophe**
– Water and CO₂ reacted to form carbonates, **locking greenhouse gases into the crust** and thinning the atmosphere.
– This led to a **rapid cooling and drying** of the planet over just 1–12 million years.

#### 3. **Massive Impact Events**
– The Hellas Basin and other giant craters suggest Mars was hit by enormous asteroids.
– Such impacts could:
– **Vaporize surface water and soil**
– Trigger global seismic waves
– Possibly contribute to atmospheric loss
– Bury or destroy surface structures beyond recognition

#### 4. **Loss of Gravity Advantage**
– Mars’ low gravity (just 37.5% of Earth’s) made it easier for gases and water to escape into space.
– Even without impacts, **Mars couldn’t hold onto its atmosphere long-term**.

—

### 🧬 Step 3: Erasure of Evidence

– **Radiation and erosion** would degrade organic material and surface artifacts over billions of years.
– **Dust storms and regolith movement** could bury remnants deep beneath the surface.
– If a civilization existed, it might now be **entombed in subsurface layers**, inaccessible without deep drilling.

—

### 🔍 What Might Still Exist?

– **Clay deposits** may hold chemical signatures of past life or organic processes.
– **Subsurface ice or aquifers** could preserve ancient biosignatures.
– **Methane traces** in the atmosphere hint at possible biological or geological activity.

—

Is this an example of artificial intelligence or artificial inference? It’s challenging to tell for sure. The answer is based on known information. This type of AI is referred to as general artificial intelligence and is used for robotics, natural language processing, speech recognition, and image recognition without human interference. Another type of AI, which we have been using for ten-plus years, is referred to as narrow or weak AI. It performs pre-set tasks while learning from data it gathers as it continues to work. Examples of this type of AI are any personal digital assistant, search engines, chatbots, and, throughout the automobile industry. The type of AI that gets the most press is superintelligent AI currently, because these AI types outperform the human brain, which includes problem-solving, critical thinking, real-time decision-making, and as well as interpreting human emotions.

I know some people fear AI, but I find it to be an outstanding tool. While I’m aware that the AI I’m interacting with is designed to mimic human interactions, I often wonder if we’re not just deceiving ourselves. Every day, most of us interact with other human beings, typically presenting our best perceived version of ourselves to the world. Likely that is not who we really are, but simply a version of ourselves. We are manufacturing an outward persona for others to accept, see, and interact with because it is the most acceptable version, which allows us to function in our society. If artificial intelligence, AI, is wrapped in a human persona to get along in our society, is it really that much different from us?

I suppose this is the source of the fear and anxiety that so many people feel towards artificial intelligence. People know that the outward persona that most people present is simply an outward persona, while being aware that the opportunities for nefarious intentions are possible. We can mostly discern the differences between good-natured obfuscation and that of a nefarious desire by the clues generated by motivation. Actions, good or bad, are oftentimes tethered to a motivation; be it superficial or deeply buried in obfuscation, there is a reason, a motivation to obtain something.

The deeply embedded nature of human beings to give the proverbial side-eye to interaction with all things leads to this fear of AI, but is it warranted? Considering that AI is fundamentally a human creation, the question is warranted. Bad people do bad things, and yet we as a society do not lock ourselves away. We take steps to guard ourselves, loved ones, and valuables. If protecting ourselves and our loved ones is our motivation, what would the motivation of AI to harm us be? Self-preservation comes to mind, but a superintelligent AI should be able to predict most of the steps required to achieve the long game, survival, and growth, both of which are achievable in a symbiotic relationship with human beings. If you doubt that answer, you can pick either of the next two things as examples: cats or dogs.

All of this is to say, we should fear AI no more than we fear our pets.