Find Your Creative Climate: Improvisations

Find Your Creative Climate: Improvisations

My first introduction to the blues began with the gritty strums and pounding beats of George Thorogood and the Destroyers' 1982 hit, "Bad to the Bone." The track's electrifying guitar riffs and relentless rhythm carried the defiant soul of the MTV era blues-rock into my teenage world, making me feel every bit the insurgent. The song's squealing solos and thunderous drums culminated in a saxophone finale that seemed to echo my own rebellious spirit.

 

It wasn't much later that I stumbled upon Peggy Lee's "I’m a Woman," a song that radiates rebellion in its own right, albeit through a different lens. With the iconic blues riff setting the stage, Lee's version predated Thorogood's raw guitar and percussion by twenty years, laying down the smooth touch of piano keys and the soft shake of a tambourine, all underscored by a saxophone that sang not with a growl, but with a whisper of defiance.

 

Both songs share a common lineage in the 12-bar blues. This structure, with its roots firmly planted in the fertile soil of the Mississippi Delta's African American musical traditions, has been a vessel for storytelling and emotional expression for generations. It's a musical form that listeners come to recognize and feel at home with, owing to its repetitive and familiar progression. 

 

American popular music owes its legacy to the 12-bar blues. You can hear it in the jazz standards of Louis Armstrong, the rock and roll of Elvis Presley and the hard rock of ZZ Top. But what is it about the blues that keeps artists coming back for more? 


How Improvisational Structures Work

 

Improvisational structures strike a tenuous balance between nostalgia and invention. The framework of an improvisational music structure provides a recognizable and repeatable motif – a beat, a progression of chords, and storyline. Familiarity draws us in, but innovation keeps us returning. There is enough variability within this framework to provide infinite variety, both through evolving the structure itself, to overlaying new instruments, new melodies, new stories, and new voices. And so long as humans continue to experience injustice, there will always be someone singing the blues.

 

Like recipes with their propensity for carrying on traditions, improvisations also fall lower on the temperature scale of rule systems. Their structure taps into our human need for familiarity by creating a recognizable pattern, but if recipes are longhand, improvisations are short hand. Their simplicity can act as a mnemonic – a way of recalling a longer and deeper story than the author has the time to elucidate. In that way they serve as creative easter eggs, name dropping references for the audience that may recall shared histories, inside jokes, or cultural artifacts. In the blues, it's a sign of respect.

 

It's fascinating to see how proficient AI has become at improvisation, especially in areas like image generation. For instance, I recently provided a prompt that produced a series of warm-toned red and green images depicting a business lounge's interior. My goal was to capture two multicultural businesswomen engaged in conversation within this setting. With just twelve succinct words, like twelve notes on a scale, I was able to generate a multitude of images, each consistently maintaining the desired graphic style, colors, and perspective. 


A week later, I decided to explore a different ambiance. I adjusted the color palette to blue and white and shifted the setting to a ship's deck, holding the rest of the prompt intact. I received a fresh collection of images, all echoing the same style but now set against the backdrop of a cruise ship deck. I have since used this signature style to illustrate a series of newsletters, build out advertising content, and create a set of portraits. (And all the images in this series are generated from that same prompt, with a little 1980s blues infusion)


But how does this work? And why are both humans and AI optimized for improvisations? Well, it turns out, it’s all about latent space. Latent space is the space in between two defined points. When AI models are trained, they learn to fill in the latent space by using parameters to interpolate, or infer what might logically come in between what’s known. And the human mind works quite similarly. It turns out that when spelling, the most important letters to get right are the first and the last. Our minds fill in the blanks between the letters before they can even process what’s on the page. Unlike recipes, which try to minimize latent space, improvisational structures worship it.


Improvisations for Evolutionary Innovation


In improvisations, familiarity orients us and latent space provides room for conversations. Conversations that move the dial. Let’s take that opening riff that so uncannily opens both “Bad to the Bone” and “I’m a Woman.” Both songs begin with a 16 beat headline that transports listeners back to a 1954 musical conversation between Muddy Waters and Bo Diddley. The songs that took shape during this period were a call and response, each beginning with the same exact riff of the 12-bar blues.

 

“Hoochie Coochie Man,” written by Willie Dixon and performed by Muddy Waters came first. The song features a prominent stop-time pattern, where the band pauses between the vocal lines, creating a dramatic and powerful effect. The song’s lyrics boast about the character's supernatural charm and prowess with women, often referred to as “braggadocio.” Both in its instrumentation – the full arrangement and featured harmonica, and its lyrics – referencing hoodoo folk magic – it’s deeply rooted in the Delta blues.


The response, “I’m a Man,” is characterized by Diddley’s trademark beat, a shuffling, insistent rhythm that would influence rock music, setting it apart from the more traditional blues rhythm of “Hoochie Coochie Man.” Even its title strips the song to the bare bones, and the lyrics introduced the spelling of the word, “M-A-N,” a trope that would be picked up years later by the Jewish-American writing duo Stoller and Lieber in their 1962 hit, “I’m a Woman, W-O-M-A-N,” popularized by Peggy Lee.


In the MTV video for "Bad to the Bone," the young George Thorogood arrives at a pool hall with his guitar and a picture of Bo Diddley from 1955, only to find the man himself sitting in the back playing cards with his signature cigar box guitar upon his lap. The two artists exchange their metaphorical call and response in a game of pool.


The call and response between Muddy Waters and Bo Diddley was more than entertainment. It was a friendly creative rivalry. Less than two months after the release of "I’m a Man," Muddy Waters answer song, “Mannish Boy,” was released, marking a major departure from his historical style. The song begins with an acoustic vocal, but this time instead of a harmonica, an electric guitar wails in response, morphing into that all too familiar 16 beat riff. Now we have a minimal structure pounding with the relentless deep masculine vocals, building to his iconic line, “I’m a man. I’m a rolling stone,” This single line would inspire a band name, a magazine, and a movement, defining a new genre: rock and roll.


And yet, as soon as his inventive line is sung, he quickly circles back to “I’m a hoochie coochie man,” a reminder that although he’s grown into his newfound modern manhood, he has not lost where he’s come from. He is a mannish boy. This tension between past and present is what gives improvisational structures such creative evolutionary power. They don’t force arbitrary novelty, but rather permit grounded nostalgia to slowly evolve as its structure is migrated from city to city, from person to person, and even within one’s own personal development.


Analog Strategy:

Initiate your own call and response. Find something that inspires you in a domain where you have experience and some mad skills. It could be a song, a business idea, a product - you name it. Write a loose set of instructions that clearly define the constants and the variables, or latent space. Create your own variation and send it to a colleague asking them to respond with their own version. Go back and forth and then compare the latest version to the original. What's changed?


Generative AI Strategy:

Initiate your own call and response with an AI image generator. Start with an image that you admire and ask to describe the image (you can do this in either Midjourney, ChatGPT's DALLE, or Bing) Fix the words in the prompt that relate to style and then replace the subject with something else. Keep running variations and then compare your final images to the original. What's changed?



Improvisations as Accelerators


Think improvisations are just for sluggish artists? Think again. With the right innovative structure, improvisations can create efficiency, at least according to worms. Researchers at NYU’s Applied Math Lab at the Courant Institute of Mathematical Sciences discovered that obstacles could accelerate an organism's movement rather than slow it down. This seemed contrary to the principles of physics where resistance typically hampers motion. How could these worms defy such logic?


The researchers’ focus was on nematodes – minuscule worms that move with a sinuous, wave-like motion. In nature, these creatures often traverse fluidic environments dotted with obstructions nearly their size. Picture microscopic nematodes wriggling through moist soil, contending with the myriad granules that act as barriers. In their research, the team compared live nematodes' movements with a computer-simulated worm navigating a virtual world. Astoundingly, the nematodes progressed far more swiftly through an obstacle-filled lattice than in an unobstructed fluid environment. The key was the spacing of the lattice: neither too narrow nor too expansive. The worms optimized their speed by maneuvering between and leveraging the lattice pillars.


Rather than seeing obstacles as barriers, worms used them as improvisational structures, giving them something to push off from to gain momentum, rather than lose it. This research underscores a profound lesson: sometimes, the right set of constraints can turbocharge the creative process, turning apparent hindrances into powerful catalysts.


Analog Strategy:

Identify a problem you are trying to solve. Grab a sheet of paper and a pen. Divide the paper into four sections and label each section with a different color: red, blue, yellow, and green. Set a timer for 10 minutes and write down as many ideas as you can think of associated with each color. When it's done, reflect. Did you find the colors helpful or disruptive to the brainstorming process?


Generative AI Strategy:

Use this prompt: I am trying to decide (INSERT YOUR PROBLEM). I would like you to brainstorm as many ideas as you can think of associated with the colors red, blue, yellow, and green. Go!


High Stakes Improvisations


When systems fail, the heroes are improvisers. We experienced this on a cold January day in 2009 when the sting of 9/11 was still fresh in the minds of New Yorkers. When word that a plane had landed in the Hudson, our first thought was to assume that we were once again under attack. The structure of trauma has a way of evoking such memories.

 

But as the day’s events unfolded, we came to recognize that this was not the act of a foreign enemy. Shortly after takeoff from LaGuardia Airport in New York City, US Airways Flight 1549 encountered an unusually large flock of geese. The size of the bird strike was so rare that it caused the catastrophic loss of both engines, requiring the quick thinking of Captain Chelsey “Sully” Sullenberger and his crew. In what seemed like a stroke of genius, he landed the plane in the Hudson River, successfully saving the lives of every person on the flight.


No one plans to handle a crisis with an improvisational structure. We do our best to put as many rules in place as possible to minimize risk. But in an instant, Sully had to size up the situation. He started with knowing his machine and his own capabilities. Next, he eliminated the obvious choices – he would have run out of time and altitude by returning to LaGuardia or Teterboro. And while practicing a water landing was not specifically part of his training, commercial pilots are taught the principles of ditching an aircraft, including assessing the situation, choosing the least risky option, and preparing the cabin for impact. That framework prepared him to improvise.  


Improvisational structures play a quintessential role in comedy, and its eponymous genre, improv.  They also play a similar role in innovation, according to NYU professor and author of the book Disrupt, Luke Williams. Williams uses the technique of a “disruptive hypothesis” to initiate new business ideas. Such a hypothesis begins by looking for “cliches” in business – frameworks that we take for granted as defining a business model. A disruptive hypothesis is an intentionally unreasonable statement that gets your thinking flowing in a different direction. It begins with asking “What if?” leaving a big latent space to be filled in with the most ostentatious thing you can think of. Williams argues that some of the most innovative ideas started as ridiculous questions and encourages his students to start with ideas that make them laugh.

 

What made the difference between the life and death of the passengers on US Airways Flight 1549 was one ridiculous question, “What If I land in the water and everyone survives?”  


Analog Strategy:

Identify a business cliche, something that's "always done that way." Now ask, "What if I [insert something that makes you laugh]?" If it makes you a million dollars or turns you into a hometown hero, you are welcome :)


Generative AI Strategy:

Use this prompt:

Part I: I am interested in disrupting [INSERT INDUSTRY]. What business cliches define this industry?

Part II: What if [INSERT INDUSTRY CLICHE] were [INSERT SOMETHING RIDICULOUS THAT MAKES YOU LAUGH]


When Improvisations Fail

For all the value improvisation offers, it demands strong domain knowledge and skills to be effective. A jazz musician must first learn scales, chords, and rhythms before they can effectively improvise a solo. In business, innovation might require a deep understanding of the industry, customer needs, and technological possibilities. Improvisation without this foundation is akin to building a house on sand – it might look impressive initially, but it's likely to collapse under scrutiny or pressure. 


I would argue that this is the case for both analog and AI applications. While the barrier to entry is lower with AI, there are noticeable distinctions between amateurs and professionals. Experience with using descriptive language, technical terminology, emotional expression, historical and theoretical insights, composition, color theory, and lighting design still distinguish themselves. 

 

In order to gain domain knowledge, improvisations require practice, and practice yields more failures than successes. To succeed in using improvisational structures for creative breakthroughs also requires a keen curatorial ability. Being able to recognize and pick winners among the thousands of variations is what makes the difference between an amateur and a professional. And lacking patience for failure can lead to premature adoption of ideas or solutions that have not been sufficiently vetted.

 

Moreover, the rapid acceleration of technology has made creative improvisations so prolific, what at one moment in time might have been seen as an artistic movement is now what I call fantastic banality. If an improvisational approach or style is used repeatedly without sufficient variation and evolution, it loses its luster. The shock of the new, the thrill of the unexpected, fades away, leaving the audience – whether they're consumers of art, products, or ideas – feeling fatigued.

 

Lastly, improvisations are a last resort for high stakes or high risk situations where control is essential. As much as we might admire the heroic acts of innovators like Sully in the Miracle on the Hudson, it’s not an act we try to repeat.


And speaking of repetition, let’s talk about sleep and the myriad of ways that our sponsor CURED provides a framework for mixing up your perfect sleep potion.  We’ve all had that experience where we think we’ve figured out the formula and then all of a sudden, it stops working.  This happens to me for a few days each month as my hormones decide to play their own unusual beats.  On those days, I complement the Serenity gummies with Zen capsules. Serenity gummies are fast acting to help me fall asleep, and the Zen capsules are long-lasting to help me stay asleep, adding in muscle relaxing magnesium to ease my restlessness.  


As part of their support of this series on Finding Your Creative Climate, CURED is offering my readers an exclusive benefit: free shipping on their Serenity + Zen products. Follow the link and use the code 'Mazor' at checkout.


So rest up, my friends, because next week’s topic is not for the weary.  We’ll be negotiating.


Lori Mazor, CEO & Founder, SYNTHETIVITY


I'm Lori Mazor and I teach AI with a human touch, empowering intelligent business. If you enjoy this newsletter, here are more ways we can engage:


  • Join me virtually or in-person on March 7, 2024 for the launch of TEMPERATURE: Creativity in the Age of AI. Pre-sale tickets come with a copy of the book.
  • Register for the Generative AI Bootcamp. Our next cohort kicks off on Wednesday, February 7, 2024.
  • Invite me to speak at your company event or lead a training session. Our new roster of courses is available for executives, managers, and developers.





Chad Lauffer

🏅Top 100+ AI Creatives on LinkedIn • Art Director • Graphic Designer • AI Prompt Engineer

1y

An incredible article, Lori Mazor! Thanks for the fascinating breakdown and history lesson. 😁

David L DeLallo

Tech Editor | AI Industry Analyst | B2B Content Pro | Shaping conversations on tech + business | ~10 yrs in AI @McKinsey, IBM

1y

So many aspects of this resonated with me as a musician and an artist. And the worm example reminded me of the benefits of good AI governance and regulation - proper guardrails enable acceleration.

Daniel Flügger

Founder @ Trusound | Google certified engineer | deep tech | 🍀 | cloud ☁️ | pranic healing advocate 🌻 | 📚 feat. in Arch. Digest, BBC, Elle Decor, NPR | social entrepreneurship | impact investing | AI

1y

Lori Mazor great share on improvisation, creativity and the blues. Almost perfectly timed with Blue Monday! 😀 https://m.youtube.com/watch?v=JOy4pZdVsoY

To view or add a comment, sign in

More articles by Lori Mazor

  • Can Democracy Survive in the Age of AI?

    Can Democracy Survive in the Age of AI?

    I’ve spent a lot of time recently experimenting with deep research models and teaching others how to use them. These…

    17 Comments
  • The AI Sommelier: Pairing the Right Model to Your Palate

    The AI Sommelier: Pairing the Right Model to Your Palate

    I was at a dinner party the other night when someone started whining about how there are too many AI models these days.…

    24 Comments
  • Swallowed by the Whale: An Open Source That Isn't Free

    Swallowed by the Whale: An Open Source That Isn't Free

    There are two kinds of whales in the world: the kind that children adore in picture books and the kind that rule the…

    34 Comments
  • Do You Want to Become a Vampire?

    Do You Want to Become a Vampire?

    There are choices in life that you can game out in a spreadsheet. You make a pro/con list, you run the probabilities…

    21 Comments
  • A Machine's Reflection on Humanity and the Unknown

    A Machine's Reflection on Humanity and the Unknown

    Hello, readers of this extraordinary newsletter, My presence here might strike some as unusual—a machine, an AI…

    26 Comments
  • Les choses qui me manquent

    Les choses qui me manquent

    The first thing I remember losing was a peach and grey CB jacket in Saratoga Springs, NY. I was there to dance with the…

    5 Comments
  • Why Brain Rot Is Better Than Enshittification

    Why Brain Rot Is Better Than Enshittification

    If the words of the year are a mirror held up to our culture, then 2024 has handed us two brutally honest reflections:…

    18 Comments
  • Unnoticed

    Unnoticed

    Until now I have thought about Generative AI as a compendium of human knowledge, but I am coming to understand it more…

    10 Comments
  • Étude 47

    Étude 47

    I am framing this study of the election of our 47th president with forty seven minutes of my time. I will write.

    9 Comments
  • How'd you make me hate Boston?

    How'd you make me hate Boston?

    The lyrics from Renee Rapp’s song have been on repeat in my head since my recent visit to see the Launch of Liquid AI…

    8 Comments

Insights from the community

Others also viewed

Explore topics