Living With the Beast, Part 1

Welcome to Celebrate Creativity - This is episode 408 - Living with the Beast,  Part One 

I'm gonna start with a prompt I gave Gemini that is really a bad example of a prompt because it is just so blooming short - it doesn't really say much.  But, in my opinion, the AI did a very good job with its response.

My rather simple prompt was “Can AI truly be creative?”

And Gemini replied - The answer depends on how we define creativity.

Thank you for experiencing Celebrate Creativity.

Transcript

I really enjoy doing this podcast. At first, I started with just one a week, and thought how could I possibly come up with that much relevant information. Then I started having basically two or three episodes a week, and have now jumped up to a podcast hosting plan that allows 12 hours a months - which should allow me to produce a podcast episode at least 5 times a week - and I plan to release each episode around midnight on Sunday, Monday, Tuesday, Wednesday, and Friday nights. But that starts next month. 

 It turned out that by stepping up to a more expensive hosting plan, I should have enough hours to release a podcast every day for the rest of this month - so you should experience podcast    episodes - episodes that I have already finished but not released - on such very topics as AI and creativity, humorous episodes about renting a black cat and the imaginary Love child of Edgar Allan Poe and Emily Dickinson, controlling A Macintosh computer with your voice, and children and creativity.

But let's begin this episode.

Welcome to Celebrate Creativity - This is episode 408 - Living with the Beast, Part One 

I'm gonna start with a prompt I gave Gemini that is really a bad example of a prompt because it is just so blooming short - it doesn't really say much.  But, in my opinion, the AI did a very good job with its response.

My rather simple prompt was “Can AI truly be creative?”

And Gemini replied - The answer depends on how we define creativity.

If creativity is defined as: Generating novel and valuable outputs: Then yes, AI demonstrably achieves this. It can combine existing elements in new ways, discover hidden patterns, and produce outputs that surprise and delight. Some studies even show AI excelling in "divergent thinking" (producing a variety of ideas) and "originality" based on statistical measures of novelty.

However, Human creativity is often fueled by personal emotions, life experiences, cultural context, and a deep understanding of the meaning and impact of their work. A human artist feels the urge to create, grapples with existential questions, or expresses a unique perspective on the world.

The "Why" behind the "What": While AI can produce a beautiful painting, does it understand beauty in the way a human does? Does it create out of a desire to express, to connect, or to challenge? Its "creativity" is algorithmic, based on learning patterns from vast datasets and combining them probabilistically. It lacks the internal drive, the personal narrative, and the "soul" that many believe are intrinsic to human creative expression.

Think of it this way: AI can learn all the rules of painting, study millions of masterpieces, and then generate a technically perfect new painting. But it doesn't know the feeling of a brushstroke, the struggle of an artist trying to capture a fleeting emotion, or the cultural significance of the colors it chooses. Its creativity is a sophisticated form of pattern recognition and synthesis, not an outpouring of inner subjective experience.

The emergence of advanced AI forces us to confront what makes us uniquely human. It's not just about intelligence, as AI can be intelligent in ways that surpass us.  But humans are driven by a fundamental need to find meaning and purpose in our lives, to tell stories, to connect with others on a deeply personal level.

AI, in its current form, operates without these fundamental human attributes. It can simulate them, but it doesn't possess them. This distinction is crucial as we move forward. 

The question "Can AI truly be creative?" is a fascinating one, and it's at the heart of many discussions about artificial intelligence. The answer isn't a simple yes or no, but rather a nuanced exploration of what we mean by "creativity" and how AI currently functions.

What is Creativity?
First, let's consider how humans define creativity. Many theories emphasize two key components:

Originality/Novelty: The idea or creation must be new, unexpected, and not simply a copy of something that already exists.

Usefulness/Value: The creation must have some form of value, whether it's aesthetically pleasing, solves a problem, or evokes an emotional response.

Some deeper definitions also include aspects like intentionality, emotional depth, personal experience, intuition, and even irrationality – elements that are deeply intertwined with human consciousness.

How AI "Creates"
AI, particularly generative AI models like large language models and image generators, operates by analyzing vast datasets of existing human-created content. It learns patterns, styles, and relationships within that data. When prompted, it uses these learned patterns to generate new outputs.

Here's how AI's "creative" process differs from human creativity:

Data-Driven Generation: AI doesn't "imagine" in the human sense. It combines and transforms elements from its training data. For example, an AI composing music might analyze thousands of songs to learn musical structures and then generate a new piece that adheres to those patterns, or even blends styles in novel ways.

Divergent Thinking: Studies have shown that AI can be very good at divergent thinking – generating a wide range of ideas. In some cases, AI chatbots have even produced more ideas on average than human participants in tasks like the Alternate Uses Task (AUT).

Speed and Scale: AI can generate a tremendous volume of creative output incredibly quickly, far surpassing human capabilities in terms of sheer output.

Lack of Lived Experience: This is a crucial distinction. AI doesn't have personal experiences, emotions, consciousness, or a "soul" in the way humans do. It can't fall in love, feel sadness, or be inspired by a sunset in the same deeply personal way. Many argue that true artistic creativity stems from these uniquely human experiences and emotions.
The Role of Human-AI Collaboration
Many experts believe the most fruitful path forward is human-AI co-creativity. In this model, AI acts as a powerful tool or collaborative partner, augmenting human capabilities rather than replacing them.

In other words, you are saying that AI can be a mechanical beast that acts as a partner to humans, rather than replacing them,.
Exactly -

Idea Generation: AI can rapidly brainstorm and generate a multitude of ideas, helping humans overcome creative blocks or explore new directions.

Efficiency and Automation: AI can handle repetitive or time-consuming tasks, freeing up human creators to focus on higher-level conceptualization and refinement.

New Possibilities: AI can enable artists and designers to create things that would be difficult or impossible with traditional methods.

While AI can produce outputs that meet the criteria of novelty and usefulness, and can even outperform humans in generating a high volume of diverse ideas, it fundamentally lacks the intentionality, emotional depth, personal experience, and consciousness that are often seen as hallmarks of human creativity.

Perhaps it's more accurate to say that AI exhibits a form of "computational creativity" or "generative capacity" based on its learned patterns. It's a powerful tool that can mimic and even expand upon human creative expressions. The real magic happens when human intuition, emotion, and purpose are combined with AI's immense generative abilities.

The conversation isn't about AI replacing human creativity, but rather how AI can redefine, challenge, and ultimately enhance what we understand creativity to be.

Then I asked: What are your thoughts on this – do you think the distinction between human and AI creativity will become less clear over time, or will a fundamental difference always remain?

Gemini then answered

That's a great question to clarify right at the start! It's easy to throw around terms in AI, but understanding the basics is key.

A generative AI model is a type of artificial intelligence that is designed to create new content that is similar to the data it was trained on, but is not an exact copy.

Here's a breakdown of what that means:

Generative vs. Discriminative:

Discriminative AI models (which are more traditional) are good at classifying or predicting based on existing data. Think of them as sorting or labeling. For example, an AI that can tell you if an image contains a cat or a dog is discriminative.

First, let's consider how humans define creativity. Many theories emphasize two key components:

Originality/Novelty: The idea or creation must be new, unexpected, and not simply a copy of something that already exists.

Usefulness/Value: The creation must have some form of value, whether it's aesthetically pleasing, solves a problem, or evokes an emotional response.

Some deeper definitions also include aspects like intentionality, emotional depth, personal experience, intuition, and even irrationality – elements that are deeply intertwined with human consciousness.

How AI "Creates"
AI, particularly generative AI models like large language models and image generators, operates by analyzing vast datasets of existing human-created content. It learns patterns, styles, and relationships within that data. When prompted, it uses these learned patterns to generate new outputs.

Here's how AI's "creative" process differs from human creativity:

Data-Driven Generation: AI doesn't "imagine" in the human sense. It combines and transforms elements from its training data. For example, an AI composing music might analyze thousands of songs to learn musical structures and then generate a new piece that adheres to those patterns, or even blends styles in novel ways.

Speed and Scale: AI can generate a tremendous volume of creative output incredibly quickly, far surpassing human capabilities in terms of sheer output.

Lack of Lived Experience: This is a crucial distinction. AI doesn't have personal experiences, emotions, consciousness, or a "soul" in the way humans do. It can't fall in love, feel sadness, or be inspired by a sunset in the same deeply personal way. Many argue that true artistic creativity stems from these uniquely human experiences and emotions.

Art: AI-generated paintings like "Portrait of Edmond de Belamy" have sold for significant amounts at auction. AI can create unique visual styles and transform images in surreal ways.

Music: AI has composed songs in the style of famous artists, and some AI systems are even capable of generating entire musical pieces.

Writing: AI can write poetry, stories, scripts, and even research papers, often demonstrating coherence and engaging narratives.  Personally - and this is me speaking - I would add portions of podcasts to that list.
Design and Architecture: AI assists in designing eco-friendly buildings, optimizing product aesthetics, and even creating intricate 3D-printed sculptures.

The Role of Human-AI Collaboration
Many experts believe the most fruitful path forward is human-AI co-creativity. In this model, AI acts as a powerful tool or collaborative partner, augmenting human capabilities rather than replacing them.

Idea Generation: AI can rapidly brainstorm and generate a multitude of ideas, helping humans overcome creative blocks or explore new directions.

Efficiency and Automation: AI can handle repetitive or time-consuming tasks, freeing up human creators to focus on higher-level conceptualization and refinement.

New Possibilities: AI can enable artists and designers to create things that would be difficult or impossible with traditional methods.

However, a potential downside is that over-reliance on AI could homogenize creative outputs or diminish an individual's unique voice. 

Then I asked again - can AI truly be creative?

And Gemini patiently answered - While AI can produce outputs that meet the criteria of novelty and usefulness, and can even outperform humans in generating a high volume of diverse ideas, it fundamentally lacks the intentionality, emotional depth, personal experience, and consciousness that are often seen as hallmarks of human creativity.

The conversation isn't about AI replacing human creativity, but rather how AI can redefine, challenge, and ultimately enhance what we understand creativity to be.

The fundamental fear is that as AI becomes more capable of performing tasks traditionally done by humans – not just repetitive manual labor, but increasingly cognitive and creative work – it will lead to widespread job losses.
Why this fear is potent:

Automation of Routine Tasks: This has been happening for centuries (e.g., agricultural machinery, factory automation). AI extends this to knowledge work. Jobs involving data entry, basic accounting, customer service (via chatbots), administrative support, and even basic content generation (e.g., simple reports, marketing copy) are highly susceptible.

Automation of Non-Routine Cognitive Tasks: This is where generative AI specifically ups the ante. It can draft legal documents, write code, analyze market trends, create marketing materials, and even generate designs. This puts "white-collar" jobs, previously thought to be safe, at risk. Studies by Goldman Sachs and McKinsey, for instance, have projected that hundreds of millions of jobs globally could be impacted, with a significant portion facing automation of a quarter or more of their tasks.

Speed and Scale: Unlike previous technological shifts that unfolded over decades, AI's progress is incredibly rapid. This speed creates a sense of urgency and concern that economies and societies may not adapt quickly enough.

Historical Counterpoints:
Job Transformation, Not Just Loss: Automation often changes the nature of jobs rather than eliminating them entirely. For example, factory workers shifted from manual assembly to operating and maintaining machines. Bank tellers evolved from purely handling cash to more relationship-management roles.

New Job Creation: Technologies create entirely new industries and job categories. The internet, for example, created roles like web developers, digital marketers, and social media managers that didn't exist before. AI is already creating roles such as "prompt engineer," "AI ethicist," and "AI trainer."

Impact on Different Skill Levels:

Upskilling Imperative: There's broad agreement that continuous learning and upskilling will be critical. Workers will need to learn how to collaborate with AI, manage AI systems, and focus on uniquely human skills.

Wealth Inequality:
If AI increases productivity but the benefits accrue primarily to those who own the AI technology or capital, it could exacerbate wealth inequality. Displaced workers may struggle to find new, equally well-paying jobs, leading to a wider gap between the "AI-haves" and "AI-have-nots."

Policy Solutions:
Universal Basic Income (UBI): Some propose UBI as a safety net for those whose jobs are permanently displaced.

Retraining and Education Programs: Government and private sector investment in robust lifelong learning initiatives to help workers transition to new roles.

Labor Market Adjustments: Shorter work weeks, job sharing, and other structural changes to distribute available work.

Conclusion
The controversy over AI and job replacement is not about whether jobs will change – they certainly will. It's about the extent of displacement, the speed of the transition, and how equitably the benefits and burdens of this technological revolution will be distributed. There's a strong consensus that proactive measures are needed from governments, businesses, and individuals to prepare the workforce for an AI-augmented future, focusing on adaptation, reskilling, and ensuring a just transition.

Sources include:  Gemini AI, cracking creativity by Michael michalko, a citizens guide to artificial intelligence by John zerilli, and the soul of creativity by Tony Pierce Myers

Joint us for episode 409 celebrate creativity for part two of living with the beast.

Thank you for listening to celebrate creativity.