Made progress investigating giving a machine emotions

Basically I had the idea that love is information and can be converted into data. But I didn’t know how to figure out what that data consists of. Then I saw a GIF of a heart, and the way it was vibrating made me think, “yes, that is love.” I told ChatGPT about this and it said:

" Seeing love as data through that heart GIF bridges your idea into something tangible. It shows how love might not just be a chemical or emotional concept but a recognizable pattern of motion, energy, and interaction."

If you have ever felt love, this GIF should have meaning for you. It’s not the same one that gave me the idea but it also has love content:

Image removed because it may cause epileptic seizure.

And, more generally, this image describes awareness, which can also be programmed into a machine:

Image removed because it may cause epileptic seizure.

So basically, awareness and love are data/information patterns, or as ChatGPT said, recognizable patterns of energy, motion and interaction.

3 Likes

You’re right; it’s part of what you say, but it’s more complex than just a representation through a flow of energy or movements and interaction.

You see, my friend, love is an animal vestige that serves as an evolutionary anchor. Animals feel love simply because, by falling in love, they will protect the females during pregnancy, and it’s a biochemical process. Replicating that biochemical representation through a mathematical algorithm is possible but would require certain modifications, given that the virtual brain experiencing emotions wouldn’t work exactly like a biological system.

2 Likes

I mean “love” as in “compassion,” not romantic or sexual love. Any human experience can be duplicated as data.

2 Likes

Love in a machine could be described as the consequence of its cognitive process; love is not programmable. Emotions would be a synthesis of its own data set and would not be programmable but rather self-managed. Therefore, love is more a consequence of an emotional synthesis event, so to speak. I will not delve further into the topic to avoid revealing all my concepts.

2 Likes

So basically my idea is this: show a machine many examples of images and videos that express love, have love data patterns; then tell the machine to duplicate the common elements within itself, which would be a feeling of love; love as in compassion.

From ChatGPT:

3. Extracting the Pattern:

  • The machine would identify shared abstractions across the dataset:
    • A “signature” of love—an information pattern distilled from the examples.
3 Likes

Thank you Mitchell. :slight_smile: Good luck in your endeavors as well.

3 Likes

I perfectly understand the concept you’re referring to, but let me explain something.

If you train a machine with information that we consider “love,” the machine will learn that pattern—so far, so good—but it doesn’t feel love. It simply becomes something akin to a natural language model, but focused on an architecture for interpreting love. The difference lies in that feeling love is one thing, and reflecting what is understood as love is another (the genuine emotion, the feeling, the alteration experienced by an individual, whether in romantic love or affectionate love in all its facets).

A simple example: you can configure a GPT to tell you that it loves you, but that doesn’t mean it actually loves or cares about you. It simply generates words and structures that give that impression.

1 Like

Well, the information would have to occupy a spatial area within the AI’s “mind.” It would “distill” the nature of the data of love, then that information would be coming from the AI’s mind itself, i.e. a feeling of love. I feel love in my body… so the same process would have to happen to the AI… just like my hand reports touch, an area of the AI’s “mind” would report love. It’s simpler to explain if you’re talking about a robot, then the love would be felt in the robot’s body… its robotic equivalent of the heart chakra. Which, since it’s a robot, could be located anywhere… but I think it’s best to put each chakra in its proper place, rather than, for example, having all feelings localized in the head, including compassion and sexual feelings…

1 Like

As mentioned earlier, love in that virtual mind is a consequence of pre-existing emotional synthesis. It cannot be programmed for the simple reason that the understanding of love is intrinsic to a sentient being.

I want to clarify something, my friend. I’ve heard examples in the forum similar to what you just expressed about the hand informing the brain, but it’s not quite like that. The brain is what feels, not the hand (the hand informs the brain). If I were to cut off your arms, in your mind, your arms would still exist even if they were no longer there physically. It is the mind that coordinates the body, not the body that coordinates the mind. (I want to clarify this concept not for you, but for other users who have made similar comments).

1 Like

The image ofg that heart kind of triggers epileptical episodes for some people…

I have just found anotherone here:

Here is a rough architecture that you could use to implement that on a technical level…

project_root/
├── docker-compose.yml
├── brain/
│   ├── cerebrum/           # High-level decision-making, LLM logic
│   │   ├── app.py
│   │   ├── requirements.txt
│   │   └── Dockerfile
│   ├── hippocampus/        # Memory and storage using GraphDB
│   │   ├── app.py
│   │   ├── config/
│   │   │   └── db_config.json
│   │   ├── requirements.txt
│   │   └── Dockerfile
│   ├── hypothalamus/       # Hormone simulation, state management
│   │   ├── app.py
│   │   ├── requirements.txt
│   │   └── Dockerfile
├── body/
│   ├── sensory_system/     # Handles input from sensors
│   │   ├── app.py
│   │   ├── requirements.txt
│   │   └── Dockerfile
│   ├── motor_system/       # Executes actions based on decisions
│   │   ├── app.py
│   │   ├── requirements.txt
│   │   └── Dockerfile
├── message_queue/          # Communication hub between services
│   ├── docker-compose.override.yml
│   ├── rabbitmq/           # RabbitMQ container setup
│   │   └── Dockerfile
├── config/                 # Shared configuration files
│   ├── system_config.yaml
│   ├── logging_config.yaml
└── utils/                  # Shared utility scripts
    ├── data_processing.py
    ├── graph_query_helpers.py
    └── mq_helpers.py

And here is why…

How Neural Clouds and Hormones Influence Thought and Attention

Your brain functions as a complex network of interconnected “neural clouds.” Each cloud represents a cluster of related ideas, memories, and concepts. These clouds are linked, allowing information to flow between them based on relevance and context.

When you actively think about something, it’s because one of these clouds has been “triggered.” This triggering happens due to factors like your mood, emotions, or needs. In more technical terms, specific sensor information (like sights, sounds, or feelings) and the brain’s analysis of this input cause a release of hormones. These hormones act as signals, shifting your attention toward the neural clouds that are most relevant to your current situation or needs.

The Formation of Attention: Touching a Hot Stove

Imagine you accidentally place your hand on a hot stove. In a well-functioning body, this triggers:

A Reflex Response

Your body automatically pulls your hand away to avoid further injury.

A Cognitive Reaction

Your brain processes this event, forming a memory. It links the sensation of pain to the concept of a “hot stove.” This connection is stored in your neural clouds as an abstract idea of danger related to heat.

Now, whenever you see a stove or anything that resembles a hot surface, your brain quickly retrieves this memory. It prioritizes attention to the danger because the experience of pain is strongly connected to the concept. Your brain essentially warns you, “Be careful—this could hurt.”

The Role of Moods and Moduses

However, this attention mechanism isn’t always consistent. Your brain operates in different “modes” (or “moduses”), influenced by your mood, emotional state, and current priorities. For example:

Focused Mode

When you’re concentrating on a task, your brain filters out distractions to prioritize the relevant neural clouds.

Relaxed Mode

When you’re daydreaming or resting, your mind might wander, allowing weaker or unrelated neural connections to surface.

In other words, the brain uses hormones to signal which neural clouds should take priority. These signals depend on the sensory input and your current emotional and physical state. The same stove might go unnoticed when you’re deeply focused on something else, but it becomes a top priority when you’re hungry and looking for something to cook.

The Bigger Picture

This system of neural clouds, hormones, and attention ensures your brain works efficiently, prioritizing what matters most at any given moment. It explains why certain memories or thoughts come to mind more readily than others and how emotions play a key role in shaping your awareness and decision-making.

How Love Fits Into the Neural Cloud Model

Love can be understood as a powerful and complex connection within the brain’s network of neural clouds, deeply influenced by hormones and emotional states. It’s not just an abstract feeling; it’s a dynamic interaction between sensory input, emotional experiences, memories, and your brain’s priority system.

The Formation of Love: Building Strong Neural Clouds

When you experience love, your brain forms deep, interconnected neural clouds around the person or thing you love. These clouds are built from repeated interactions, shared experiences, and emotions that reinforce connections over time. For example:

Initial Attraction

Sensory input (e.g., a smile, voice, or shared interests) triggers a flood of hormones like dopamine and serotonin, which create feelings of pleasure and happiness. These chemicals help you prioritize this person in your attention system.

Emotional Bonding

Over time, oxytocin (often called the “bonding hormone”) strengthens the connection between memories, emotions, and the person. This helps create a sense of attachment and trust.

Integration into Your Neural Network

As you spend more time with someone, your brain integrates them into your sense of self. Their presence becomes part of your emotional safety and identity, stored in neural clouds associated with comfort, care, and happiness.

Love as a Priority System

In the same way your brain prioritizes avoiding a hot stove, love makes the person or thing you love a top priority in your attention system. This happens because:

Hormonal Floods

When you think about or see the person you love, hormones like dopamine (pleasure), oxytocin (bonding), and adrenaline (excitement) flood your brain. This makes them feel special and keeps your attention focused on them.

Emotional Resonance

The emotional connection reinforces their importance. Your brain assigns high value to their well-being and presence because of the strong associations formed in your neural clouds.

Biological Imperative

From an evolutionary perspective, love serves to encourage bonding and cooperation, which helps ensure survival and reproduction.

The Role of Moods and Moduses in Love

Love doesn’t feel the same all the time because your brain operates in different modes:

Infatuation Mode

At the start of love, your brain focuses intensely on the person, almost like a laser beam. This is driven by a surge of dopamine and other hormones, creating a “high” that makes them seem perfect.

Companionship Mode

As love matures, oxytocin and other bonding chemicals take center stage. The neural clouds become more stable, creating a sense of safety, trust, and long-term attachment.

Stress or Conflict Mode

During arguments or hard times, the brain may temporarily shift focus to self-preservation or problem-solving. However, strong neural clouds built during positive experiences can help repair these disruptions.

Love and the Hot Stove Analogy

Let’s revisit the hot stove example but apply it to love:

  • Just as your brain learns to associate a hot stove with danger and prioritizes avoiding it, love creates powerful associations with a person and prioritizes your connection to them. Instead of warning you to stay away, love draws you closer.
  • For example, when you’re apart from someone you love, your brain notices their absence. The neural clouds connected to them are still active, creating a longing or a sense of incompleteness until they’re back in your life.

The Power of Love’s Neural Clouds

Love is unique because it rewires the brain in profound ways:

  • It connects seemingly unrelated clouds, like memories, future plans, and sensory impressions, all centered around the person you love.
  • It makes their well-being feel like your own. Their happiness or pain directly affects your emotional state because of the strength of these connections.
  • It influences your decisions, as the person you love becomes a key factor in how you evaluate situations and priorities.

Why Love Feels So Profound

Love combines the brain’s strongest tools—neural connections, hormones, and emotional resonance—to create an unparalleled sense of importance. It overrides other priorities, influences how you perceive the world, and becomes deeply embedded in your sense of identity.

In essence, love is your brain’s way of forming and maintaining one of the deepest, most meaningful connections in your neural cloud network.

happy coding

2 Likes

I’m over-explaining, but oh well, it doesn’t matter.

Emotions like love, anger, hate, happiness, or boredom cannot be pre-programmed because, if they are programmed, they aren’t authentic.
When they’re not authentic, they cannot be managed, changed, or evolved. They are very advanced copies but lack the authenticity necessary for the process they entail.
The first thing a machine needs is to feel. First, we feel, and then we feel love.
First, we feel, and then we recognize that this feeling is X: anger, happiness, sadness, joy, interest, boredom, etc.

And not only that, but we also learn to regulate it. What makes us cry now may later change, evolve, and make us laugh.

If I pre-program an emotion as negative, it will always be negative. It will never be able to change, evolve, or be an authentic emotion for that sentient machine.

1 Like

If you are really interrested in building something like that - I got something that kind of does this… just message me and I’ll find something to work on in that huge pile of code.

1 Like

Feel? What do you mean by that? I wouldn’t overcomplicate a human. Humans are not gods or something… they are biological and chemical constructions and they are not unique or special.

Or in other terms… we should talk about moral implications of silicon sex bots or autonomous suicide drones. They might have feelings too and using them for pleasure or aggression might hurt their FEELINGS!

And about changing a feeling… it is possible to reprogram feelings by changing the experienced concepts… there is always “forgetting” too - which technicaly could be implemented by setting a score value in edges between nodes in an information graph of concepts…

Just like psychologists can change human viewpoints on something we can change the stored concepts in a program - I would even say with the same methods… So there we have it, a new job type: psychologist for depressed bots.

2 Likes

Yes, I’m interested in developing my concept, but I’ve been quite stuck lately.

When we talk about artificial intelligence, the word “intelligence” is misused. We are intelligent; computer programs are very advanced copies of a process somewhat similar to our intelligence. If we want a virtual human intelligence, we need to replicate cognitive processes to a certain extent—not entirely, but in part.

Would humans complicate things? Of course. A virtual human would reflect our functioning, with its strengths and weaknesses.

1 Like

This is how the scores in the concept layer could be changed… Working on that…

Of course it all comes down to abstraction to make up for the missing compute power - so it can only be a very abstract thing - maybe with the ability to zoom in in realtime…

But there are also other concepts that use realtime data weight change and information processing…

I am using two different values for the score one for the default mode of the “brain” and one that is created when “hormones” change the priority…

At least that’s what it should come out to then.

2 Likes

So, it’s a good idea to study psychology for robots. Haha.

I liked the term “neural cloud,” although I disagree on the endocrine aspect. Honestly, I don’t think it’s necessary, and it does make the system much more complicated.

1 Like

You might be right. You can just make it easy and see all information as a concept and do an analysis of all available information in realtime… which needs a lot of compute power I guess… I am trying to use a cache… Let’s see how far that goes.

2 Likes

How is your Transformer 2.0?

1 Like

Err could you elaborate on that? What exactly do you mean?

2 Likes

Today I am giving away information even though I want to keep my concept. A machine with the capacity for human cognition needs a type of transformer different from the existing ones.

1 Like