This is the second in the three-part series of posts examining how speechwriters can use generative A.I. tools like ChatGPT.

In Part 1 of this series, I looked at ways ChatGPT could help write a best man’s speech for a wedding. I’ll now turn to the type of corporate speechwriting I did for over 15 years at technology companies in Silicon Valley: Sun Microsystems, H-P, and Cisco.

Perhaps my most succinct, if offbeat, summary of what that job involved was my 2008 post, The Medieval Speechwriter. I’ll use it as a framework to show how the speechwriting of the 21st century (explained in terms of the 14th) works using generative A.I. tools.

Caution: Please remember that you should never take A.I. output at face value — it can suffer from ‘hallucinations’ and generate misinformation. Google’s Bard program famously supplied incorrect data about which satellite first took pictures of a planet outside the Earth’s solar system. Result: Alphabet Inc (Google’s parent company) lost $100 billion in market value.

The speechwriter employing these tools should double-check everything. Also, I’m not suggesting you take any of the content and cut and paste it into a script. Use common sense and make this your point of departure.

The Motley Fool

I wrote that the role of the corporate speechwriter, like the Fool in the Medieval court, includes looking at issues and topics with fresh eyes. Indeed, Justina Chen, in an awesome presentation on corporate speechwriting at the 2012 Ragan conference notes

Don’t be afraid of discussing failures, the crucible moments in the company history is where we see character emerge. Speechwriters can research the times the company has failed and show the lessons, the backbone, and the spine of the organization.

Given the current limitation that ChatGPT has only ‘learned’ from data prior to September 2021, it might be necessary to switch to Bing for current information. However, for a quick list of the risks of failure in a flagship product that an executive might wish to acknowledge and address, take this example:

PROMPT: Act as an industry analyst. List five pros and cons about Meta’s VR headsets

As an industry analyst, I can provide you with a list of five pros and cons about Meta’s VR headsets (formerly known as Facebook’s Oculus VR headsets). Please note that the information provided is based on my knowledge cutoff in September 2021, and there may have been updates or changes since then. Here are the pros and cons:

Pros of Meta’s VR headsets:

1. Immersive Virtual Reality Experience: Meta’s VR headsets offer a highly immersive virtual reality experience, allowing users to feel fully immersed in a virtual environment. The headsets typically have high-resolution displays and a wide field of view, enhancing the visual experience.
2. Strong Content Ecosystem: Meta has built a robust content ecosystem for its VR headsets. The Oculus Store provides a wide range of VR games, applications, and experiences, offering users a diverse selection of content to enjoy on their headsets.
3. Growing Developer Community: Meta has cultivated a strong developer community around its VR platform. This has resulted in a steady stream of innovative and engaging content being developed for the platform, including games, educational apps, and creative experiences.
4. Advanced Tracking and Controllers: Meta’s VR headsets often come with advanced tracking technology and controllers. These tracking systems allow for precise and accurate movement tracking, enhancing the overall immersion and user experience. The controllers provide a natural and intuitive way to interact with the virtual environment.
5. Cross-Platform Compatibility: Meta’s VR headsets are compatible with various platforms, including Windows and, in some cases, even standalone operation. This compatibility broadens the potential user base and allows for flexibility in using the headsets with different devices.

Cons of Meta’s VR headsets:

1. High Cost: VR headsets, including Meta’s offerings, can be quite expensive. The cost of the hardware itself, along with additional accessories and games, can make VR an investment that is out of reach for some consumers.
2. Limited Mobility: Most of Meta’s VR headsets require being tethered to a powerful gaming PC, limiting the mobility and portability of the experience. While there are standalone options available, they may have limitations in terms of processing power and graphical capabilities.
3. Potential Discomfort: Extended use of VR headsets can cause discomfort for some users. Factors like motion sickness, eye strain, and discomfort from wearing the headset for extended periods may affect certain individuals.
4. Limited Field of View: Although Meta’s VR headsets offer a wide field of view compared to earlier iterations, some users may still find the field of view to be restrictive. This limitation can impact the overall sense of immersion and visual experience.
5. Privacy and Data Concerns: Meta’s VR headsets, like many other connected devices, raise concerns about privacy and data collection. The company collects user data to provide personalized experiences and targeted advertising, which may raise privacy concerns for some users.

Please note that as the VR industry evolves rapidly, it is essential to stay updated with the latest information and product developments to have the most accurate and relevant analysis.

Let’s ask Bing for an update.

PROMPT: List the advantages that Apple’s VR headset has over Meta’s VR headset.

Apple’s VR headset has not been released yet, so it is difficult to compare the two headsets. However, based on what we know so far, here are some advantages that Apple’s VR headset has over Meta’s VR headset:

1. Apple’s VR headset is expected to have a higher resolution than Meta’s VR headset ¹.
2. Apple’s VR headset is expected to have a higher refresh rate than Meta’s VR headset ¹.
3. Apple’s VR headset is expected to have a longer battery life than Meta’s VR headset ¹.
4. Apple’s VR headset is expected to be lighter than Meta’s VR headset ¹.
5. Apple’s VR headset is expected to have a more comfortable design than Meta’s VR headset ¹.

That being said, it is important to note that these are just expectations and we will have to wait for the release of Apple’s VR headset to see how it compares to Meta’s VR headset.

The Ploughman

In 2008 I wrote: Tedious but necessary work plays a large part in speechwriting. Doing research. Fact-checking. Ploughing through the background papers which spew from Subject Matter Experts like weeds sprouting on a April morning after a few sweet showers.

Back in the day, we used the Google machine for much of this work. That, plus asking SMEs to share their research, will occupy much of the time of a corporate speechwriter. A significant advantage of generative A.I. tools is feeding in documents and extracting a summary.

As an example, the scientific paper A Review on Kombucha Tea-Microbiology, Composition, Fermentation, Beneficial Effects,Toxicity, and Tea Fungus on the KBI website has a section on ‘The Chemical Composition of Kombucha Tea’ (p 540-542). Pasting this text into ChatGPT and asking for a 10-point summary generates:

The main points covered in the paper “Chemical composition of kombucha tea” can be summarized as follows:

1. Chemical analysis of kombucha tea reveals the presence of various organic acids, including acetic, gluconic, glucuronic, citric, L-lactic, malic, tartaric, malonic, oxalic, succinic, pyruvic, usnic acids, among others.
2. Kombucha tea also contains sugars such as sucrose, glucose, and fructose, as well as vitamins B1, B2, B6, B12, and C.
3. The beverage contains 14 amino acids, biogenic amines, purines, pigments, lipids, proteins, hydrolytic enzymes, ethanol, antibiotically active substances, carbon dioxide, phenol, tea polyphenols, minerals, anions, DSL (insufficiently known products of yeast and bacterial metabolites).
4. The composition of kombucha can vary based on factors such as the duration of fermentation, the content of black tea, and the parameters of fermentation (temperature, time, initial content of sucrose and black tea), as well as the composition of the kombucha culture.
5. The metabolic activities of yeasts and bacteria in kombucha involve the utilization of different substrates and the production of various metabolites, including acetic acid, gluconic acid, glucuronic acid, lactic acid, and citric acid.
6. The content of acetic acid in kombucha increases over time during fermentation and then gradually decreases.
7. Glucuronic acid and gluconic acid are major organic acids produced during the kombucha fermentation process. The concentrations of these acids can vary based on the duration of fermentation and the composition of the substrate.
8. L-lactic acid is detected and determined in kombucha, although it is not a characteristic compound of the traditional beverage. Its concentration can vary based on the type of tea used for fermentation.
9. Sucrose, the most common carbon source in kombucha fermentation, is not completely fermented during the process, with a portion remaining unfermented.
10. Kombucha tea contains a variety of minerals, including manganese, iron, nickel, copper, zinc, lead, cobalt, chromium, and cadmium, although the concentrations of these minerals can vary.

These points highlight the diverse chemical composition of kombucha tea, which is influenced by various factors and the metabolic activities of yeasts and bacteria during fermentation.

More impressive, perhaps, is the ability of ChatGPT to perform similar time-saving tasks for the corporate speechwriter ploughing through source material in other languages. Take Section 3.2 from this French paper on the Relationship Between Comprehension Ability and Verbal Areas of the Neocortex: A Brief Study. Google Translate gave me the title in English. But pasting section 3.2 ‘Relation entre intelligence et facilité d’expression‘ (in French) and asking for the 10 main points (in English) generates:

Here are the 10 main points of the paper:

1. The concept that each person possesses a fixed and genetically determined degree of intelligence is completely false.
2. The degree of ease with which individuals understand things varies from person to person and is not fixed or genetically determined.
3. The degree of intelligence can vary throughout a person’s life based on the level of verbal expression and the habit of questioning one’s own conclusions.
4. Verbal expression and the willingness to question one’s own conclusions can increase or decrease based on intellectual activities practiced.
5. Engaging in activities such as constant reflection, problem-solving, reading, and speaking on various subjects can lead to a deeper understanding.
6. Starting to engage in intellectual activities that were previously not practiced can increase one’s understanding.
7. If a person stops engaging in these activities, their overall understanding capacity gradually decreases.
8. Changes in intellectual state occur progressively and are often unnoticed, similar to the growth process of a tree.
9. Human thought processes can be categorized into two forms: thinking through association of images and thinking through association of words.
10. The use of language enables us to describe and understand sensory perceptions, leading to verbal thought processes and increased comprehension abilities.

This took less than no time, a clear advantage for speechwriters who need to research material outside their monoglot backyard.

The Miller

As mentioned above, generative A.I. can condense and summarize the most turgid information.

The Alchemist

This is the heart of the matter. Speechwriters like Murray celebrate the ‘mystique’ of the human touch, where, as I wrote in 2008, “you’ll take their words and sit in your room all night spinning them into gold. And next morning no-one will know how you did it.”

Undoubtedly, great speechwriters will beat generative A.I. at this part of the game — at least for now. But when machines can win at chess and Go, don’t bet against them in the long term. And quite frankly, most corporate executives who writers work long hours to support couldn’t care less about sparkling rhetoric. While Justina Chen claims that “It’s important that a leader be a good storyteller, but equally important that the leader embody that story,” understand she was working with people at the top of the executive hierarchy. For every Steve Jobs keynote, there’s a hundred less able executives unwilling to engage with a writer to the level needed to create compelling content.

I’ll just say that any speechwriter searching for that perfect turn of phrase or summary story should be able to use generative A.I. to grease the creative wheels. I’ve had fun in this regard playing with A.I.’s take on literature and had it suggest a screenplay. Within the confines of most corporate presentations, it should be able to offer more than sufficiently good creative ideas and turns of phrase.

Suggestion: Take the transcript of a past talk by your executive and ask ChatGPT to revise a first draft in that voice.

The Monk

The Medieval analogy I chose to reflect the intense effort corporate speechwriters put into creating PowerPoint slides was the work of monks in the Scriptorium. There’s a whole industry designed to help develop effective presentations for ‘visual storytelling.’ Obviously, a text-based generative A.I. tool like ChatGPT can only help with this in a limited way.

I’ve experimented with MidJourney to produce interesting, creative images which could be included in slides (e.g. the images at the top of this series). But, to be honest, most companies employ graphics people for their logos, product pictures, and other branded images.

However, there’s a growing list of A.I. tools specifically designed to help with slides. These claim to be able to create designs for you in a jiffy, organize your slides to help you tell a story, and include design templates.

Canva has text-to-image generation capabilities specifically tailored to presentations.

Professional designers can now access Adobe Photoshop’s Firefly with generative fill capability.

The Wandering Minstrel

I concluded my trip through Medieval speechwriting with advice about making a living as a freelance speechwriter. The need to work fast and stay at the top of your game demands, for today’s freelancer, embracing generative A.I. as a competitive advantage.

Next: The political speechwriter.

This is the first in the three-part series of posts examining how speechwriters can use generative A.I. tools like ChatGPT.

There are concerns in the professional speech writing community about the impact of generative A.I. tools like ChatGPT on the profession. David Murray, the well-respected head of the Professional Speechwriters Association, writes, “For a real communicator, having a machine write a first draft is not a shortcut, it’s a short circuit.”

I disagree.

I believe that “You won’t lose your job to AI, you’ll lose your job to someone using AI.”

There are many ways to employ these tools. In this three-part series, I’ll look into different areas of speech writing and see where the savvy communicator can make use of generative A.I. — just as most have, over the last few decades, made use of the word processor over the typewriter, spell check over Merriam-Webster, and — dare I say it — Grammarly over Microsoft Word suggestions.

Let’s start with the lowest common denominator – the Best Man’s speech at a wedding.

There are plenty of “speech consultants” offering their services for $100 on up to script a best man speech. Faced with the challenge of writing a humorous, heartwarming, well-structured speech, many who have never addressed a room full of their best friends’ new relatives might turn to a pro to help them out.

Here’s where ChatGPT can help. At no cost.

The raw material

This wedding announcement is adapted from the New York Times. Names and locations have been changed. A best man will, of course, have all the background about the newlyweds he is to honor and can replace this text with relevant detail.

“When John Stuart Mill and Mary Andrea Wilson were first introduced in January 2011 at a dorm party in their junior year at Tufts, the sparks didn’t exactly fly.

A few weeks before the start of their senior year, the pair, who were casual acquaintances,  found themselves on an abandoned campus training together to be residential staff members for Tufts’s on-campus housing.

“We became very, very close friends our senior year because we were trapped on campus together,” said Ms. Wilson, as Mr. Mill laughed in the background.

While neither Mr. Mill, 31, nor Ms. Wilson, 30, can remember the moment their friendship blossomed into something more, they shared their first date on Valentine’s Day 2012.

“It was sort of a friendship that took us a while to figure out that it was more than that,” Mr. Mill said.

From Medford, Mass., they made their way to Boston for pizza at Luigi’s, an Italian restaurant. Mr. Mill gave Ms. Wilson roses, one of which she dried and still has.

“We were 21 years old, so we really did not go out much at that point,” she said. “It was a big deal.”

The relationship quickly turned serious. “After the first date, it was very clear,” he said. “I think the thing that was scary for us was that we knew we were graduating a few months after that.”

By this time, the couple had already accepted jobs on opposite coasts: Mr. Mill as a management consultant with the Boston Consulting Group in San Francisco, and Ms. Wilson as a White House intern for former Vice President Joseph R. Biden Jr.’s policy staff in Washington.

Each graduated with honors — Mr. Mill with a degree in economics and Ms. Wilson in history.

For about a year the couple made long distance work; both booked weekend flights to see each other.

In late 2014, Mr. Mill was offered a position in Manhattan, which Ms. Wilson saw as an opportunity to work on Hillary Clinton’s then-unannounced presidential campaign.

The couple moved in to their first apartment together in the West Village in February 2015. Shortly after, Ms. Wilson began working in Brooklyn as an aide on Mrs. Clinton’s presidential campaign.

Their idyllic life in Manhattan was cut short on election night. Mrs. Clinton was expected to deliver her victory speech in the Jacob K. Javits Convention Center. Instead what conspired was a teary run-in with Cher backstage and a lamenting walk home along the Hudson River during sunrise.

As difficult as the loss was for her, the walk home after solidified how foundational her relationship with Mr. Mill was in her life. “I think it really forced us to clarify and crystallize what we both wanted because the path that I think we thought the country and us as a couple were on changed overnight,” she said.

Before the election night, Ms. Wilson applied to law school. Her father, John David Wilson, is a retired lawyer, and her mother, Gwendolyn Adshead, is a managing partner of Fox McElgun Griffin & Adshead.

Mr. Mill was also considering graduate school, but the most important factor was to avoid any long distance.

The couple moved back to Boston, and in September 2017 started their lives as graduate students at the place where it all began, Tufts.

Mr. Mill received an M.B.A. in 2019 from Tufts, and Ms. Wilson received a law degree this past June, also from Tufts.

In December 2018, Mr. Mill proposed to Ms. Wilson during a hike to a secluded beach while on vacation in Hawaii.

The couple had been planning a barn wedding in Sonoma, Calif. for June 2020, with 175 guests, but instead had a small ceremony with immediate family at Ms. Wilson’s father’s home on Oct. 24 (which was also where they spent the first five months of the pandemic). Dr. Alfus Mill, the groom’s father and a Universal Life minister, officiated. The groom’s mother, Dr. Sandra David, was also in attendance.

Ms. Wilson, who will hyphenate her last name, is a law clerk to Angus Fitch, an associate justice of the Massachusetts Supreme Court. Mr. Mill is a founder of a money laundering and fraud prevention software company starting later this year.

“Being able to move forward and think about building a life with John was a really special moment,” Ms. Wilson said. “It’s cliché to say ‘you married your best friend,’ but I really did.”

The prompt

Act as a speech writer. Create a 600-word speech for the best man to deliver at a wedding reception. The content must be funny, with some appropriate jokes, and filled with heartwarming sentiments, flow smoothly, with a classy introduction and closing remarks. Base it on the wedding of these two people: [INSERT TEXT ABOVE.]

ChatGPT output

Ladies and gentlemen, distinguished guests, and of course, the stars of the evening, John and Mary! Good evening and thank you all for being here to celebrate this wonderful occasion. As the best man, I have the great honor of standing before you today to share some thoughts about this incredible couple.

When John and Mary first crossed paths at that dorm party in 2011, I’m pretty sure they didn’t imagine that they would end up standing here today. You see, the sparks didn’t exactly fly right away. But little did they know that fate had bigger plans for them.

It was during their senior year at Tufts that their friendship truly blossomed. As they found themselves trapped on campus together, they became very, very close friends. And let’s be honest, folks, what better way to bond than being stuck in the same place? I can only imagine the adventures and shenanigans they must have gotten into during that time.

Now, it took them a while to realize that their friendship was something more than just that. But on that fateful Valentine’s Day in 2012, they decided to take the plunge and go on their first date. And let me tell you, it was a big deal! They went to Luigi’s, an Italian restaurant in Boston, and John, being the romantic guy he is, gave Mary roses. And guess what? She actually dried one of those roses and still has it today. Now, that’s true love, folks.

Their relationship quickly became serious, even though they knew that graduation was just a few months away. But being the determined couple they are, they didn’t let distance get in their way. With John in San Francisco and Mary in Washington, they made it work with weekend flights and endless phone calls. And let’s not forget the honors they both achieved in their respective fields—John in economics and Mary in history. Talk about a power couple!

But their journey took an unexpected turn on that infamous election night. Instead of celebrating Mrs. Clinton’s victory, Mary found herself taking a walk home along the Hudson River, reflecting on the changes that occurred. It was a difficult time, but it also solidified the importance of their relationship. It made them realize that no matter what challenges they faced, their love and commitment to each other would always prevail.

And prevail it did! They moved back to Boston, where it all began, and embarked on their graduate studies at Tufts. John earned his M.B.A., while Mary recently graduated with a law degree. Talk about an intellectual power couple!

Then came that magical moment in December 2018, during a hike on a secluded beach in Hawaii, when John got down on one knee and proposed to Mary. And of course, she said yes! It was a testament to their deep love and commitment to each other.

Their wedding plans may have changed due to unforeseen circumstances, but nothing could diminish the joy and happiness that radiates on this special day when they’ve exchanged their vows. Surrounded by their loved ones, they’ve embarked on this new chapter of their lives.

Now, as they continue their journey together, Mary as a law clerk and John as a founder of a remarkable software company, I have no doubt that their bond will only grow stronger. They truly epitomize the saying, “You married your best friend.” And Mary, I must say, John is one lucky guy!

So, ladies and gentlemen, let’s raise our glasses to John and Mary, a couple who proves that love knows no bounds. May your life together be filled with laughter, joy, and endless adventures. Here’s to a future filled with love, success, and happiness! Cheers!

What would Jerry say?

You might not use the material, but what would Jerry Seinfeld say?

Prompt: Rewrite the speech in the manner of Jerry Seinfeld

Ladies and gentlemen, what’s the deal with weddings, am I right? I mean, here we are, celebrating the union of John and Mary. When they first met at that dorm party, let’s just say the sparks didn’t exactly light up the room. But hey, who needs fireworks when you’ve got an abandoned campus to train as residential staff members, right?

Now, John and Mary became close friends during their senior year because, well, they were trapped on campus together. I can only imagine the wild adventures and the crazy stories they must have from that time. It’s like a sitcom waiting to happen, trapped with your future spouse. I smell a potential spin-off!

But let’s fast forward to their first date. It took them a while to figure out that their friendship was something more. I mean, they were 21 years old, and going out for pizza at Luigi’s in Boston was a big deal! And hey, John, giving Mary roses on that date, smooth move! She even dried one of those roses and still keeps it today. That’s some real commitment, folks.

Now, after graduation, things got real. John landed a management consulting gig in San Francisco, while Mary became a White House intern in Washington. Talk about a long-distance relationship! But they made it work, with weekend flights and probably more Skype calls than they’d care to admit. I mean, they were basically running their own airline company with all those flights!

Then, destiny threw them a curveball on election night. Instead of celebrating, Mary found herself taking a walk home along the Hudson River. It was a tough time, but it made them realize what truly mattered. You know, I’ve always said that politics can change overnight, but love, that’s the real deal. It’s like a rock-solid foundation that keeps you grounded, even when Cher makes you cry backstage.

But let’s not forget their academic achievements. John acing economics and Mary acing history, they’re like a power couple of knowledge. It’s like having a walking encyclopedia for a husband and wife. I hope they never have to compete on Jeopardy, because they’d clean house!

And speaking of clean, let’s not forget that hike in Hawaii. John got down on one knee and popped the big question. I bet the view was amazing, but let’s be honest, it’s hard to compete with a Hawaiian sunset. Mary, you made the right choice. John is a catch, and I’m not talking about fish in Hawaii. He’s the real deal, and together, you guys are unstoppable.

So, here we are, celebrating their wedding, even if it didn’t go exactly as planned. But hey, that’s life. Sometimes things don’t go according to plan, but you make the best of it. And from what I’ve seen, John and Mary know how to roll with the punches.

So, let’s raise our glasses and toast to John and Mary, a couple who proves that friendship can turn into something amazing. May your journey together be filled with laughter, love, and lots of funny stories to tell. Congratulations, you crazy kids!

Hearken, fair gentlefolk, to this tale I bring, 
Of John Stuart Mill and Mary Wilson true, 
Whose paths entwined, like birds on lofty wing, 
In Tufts' fair halls where fate did them pursue.

A friendship forged, their hearts began to soar, 
As seniors trapped on campus, close they grew, 
And on that eve, love's bloom did they explore, 
Unveiling secrets that their souls well knew.

From Medford's lands to Boston's vibrant streets, 
They ventured forth, their love to celebrate, 
In Luigi's realm, where passion's flame meets, 
And roses gifted sealed their destined fate.

Through trials, distance, hardships they did tread, 
Their love prevailed, and now they wed.

An alternative prompt

Perhaps you don’t want a verbatim script, or a corny Shakespearean sonnet. Ask for an outline and script your own remarks.

Act as a speechwriter. Develop an outline of speaking points for the best man to deliver at a wedding reception. The content must be funny, with some appropriate jokes, and filled with heartwarming sentiment, flow smoothly, with a classy introduction and closing remarks. Base it on the wedding of these two people: [INSERT TEXT ABOVE.]

ChatGPT output

• I. Introduction
  • A. Greeting and acknowledgement of guests
  • B. Thanking the couple for the honor of being the best man
• II. Humorous Opening
  • A. Joke or funny anecdote about the couple’s initial meeting at the dorm party
  • B. Highlight the amusing situation of being trapped on campus together and becoming close friends
• III. Heartwarming Sentiment
  • A. Express admiration for the couple’s journey from friends to lovers
  • B. Emphasize the significance of their first date on Valentine’s Day and its impact on their relationship
  • C. Mention the challenges they faced with graduation and long-distance, but how they made it work
• IV. Funny Story
  • A. Share a lighthearted anecdote about their time living together in Manhattan and the election night incident
  • B. Add a touch of humor by recounting her encounter with Cher and the walk home along the Hudson River
• V. Classy Transition
  • A. Transition to the couple’s career and educational achievements
  • B. Highlight their dedication and commitment to their respective fields
• VI. Heartfelt Proposal
  • A. Narrate the romantic proposal during the secluded beach hike in Hawaii
  • B. Express joy and happiness for the couple’s decision to spend their lives together
• VII. Closing Remarks
  • A. Congratulate the couple on their beautiful wedding ceremony
  • B. Offer well wishes for their future as they build a life together
  • C. Raise a toast to the newlyweds and their enduring love and friendship
• VIII. Final Humorous Note
  • A. End the speech with a light-hearted joke or witty remark about the couple’s future endeavors

Note: Remember to tailor the speech according to your relationship with the couple and the specific dynamics of the wedding reception. Personalize it with additional anecdotes and jokes if appropriate.

Prompt engineering

These examples of speeches generated by ChatGPT offer an introduction to the tool. More sophisticated results can be achieved by using more elaborate prompts. Some term this ‘prompt engineering’.

Here are some examples (try them to see the results)

• Act as a speechwriter. Create an outline for a speech honoring a teacher who is retiring after many years of service using the structure of the Hero’s Journey.
• Act as a speechwriter. Create an outline in terms of the problems and solutions to help convince the audience of the need to change.
• Act as a speechwriter. Outline a eulogy for a young person who was killed in a tragic accident.
• Act as a speechwriter. Outline a eulogy for an elderly man who lived to be 102 years old and passed away peacefully.

Final word

It’s only fair to give ChatGPT the final word.

Prompt: List 10 ways in which ChatGPT can assist a professional speechwriter.

1. Generating Speech Outlines: ChatGPT can help a professional speechwriter by providing structured outlines for speeches on various topics, saving time and offering a solid foundation to build upon.
2. Brainstorming Ideas: ChatGPT can engage in creative conversations, suggesting unique angles, themes, or anecdotes that can enhance the speech and make it more engaging for the audience.
3. Providing Statistical Data and Facts: ChatGPT can retrieve relevant statistical data, facts, and figures to support the arguments or claims made in a speech, ensuring that the information presented is accurate and reliable.
4. Offering Language and Style Suggestions: ChatGPT can provide assistance in refining the language and style of the speech, suggesting alternative phrases, improving sentence structure, or enhancing the overall flow and coherence of the content.
5. Fact-Checking and Research: ChatGPT can help verify information and conduct research on specific topics to ensure the accuracy and credibility of the content included in the speech.
6. Crafting Compelling Introductions and Conclusions: ChatGPT can help in formulating attention-grabbing openings and impactful closing statements that leave a lasting impression on the audience.
7. Providing Quotes and Inspirational Messages: ChatGPT can offer a selection of quotes, inspirational messages, or anecdotes to include in the speech, adding depth and emotional resonance to the content.
8. Offering Cultural References and Examples: ChatGPT can provide cultural references, historical events, or contemporary examples that can be integrated into the speech to resonate with the audience and make it more relatable.
9. Tailoring Speeches for Specific Audiences: ChatGPT can assist in customizing speeches for different target audiences, considering their demographics, interests, or specific knowledge gaps, ensuring the content is relevant and engaging.
10. Reviewing and Editing: ChatGPT can proofread and suggest edits to refine the speech, checking for grammatical errors, clarity, and consistency, ultimately enhancing the overall quality of the final product.

Next: How generative A.I. can help the corporate speechwriter.

On a recent episode of the charming Start the Week BBC Radio 4 program we heard Frances Wilson, Salman Rushdie and Simon Armitage discuss the rehabilitation of DH Lawrence and the relationship between the writer and their work. In the closing minutes (approx 37:00) host Andrew Marr asks Armitage about an essay that claims Bob Dylan’s lyrics “don’t really hold up as great poetry, you have to listen to the songs in order to understand the interleaving of the music and the words….can you tell us what you think doesn’t work on the page? Is it repetition, is it sloppiness of the language, evasiveness?”

Armitage responds that in a poem “everything that happens in a poem has to happen within the text, in silence really…it has to be done with the alphabet.” Whereas songwriters can “write the lyric and even ‘la-la-la’ might sound very good if you put it to the right chord progression. It can be utterly transcendent if it’s working well with the music and if you’re wearing the right cowboy boots and if you’ve got the right voice.”

He reports taking Dylan lyrics into his poetry class where younger students (unfamiliar with the tune) might appreciate the poetic aspects, nevertheless see cheesy rhymes, repetition, tautology and other elements “that you can’t get away with on the printed page.”

Speechwriters as songwriters

In many ways Armitage is stating the obvious. Songs are created to be listened to, not read (as rewarding as a close reading of Dylan’s lyrics can be). As with great songs, so with great speeches.

Speechwriting authorities Nancy Duarte and Bob Lehrman drive home the message that speeches (like songs) need to be written for the ear not, (as are poems) for the eye.

Sparklines

Nancy Duarte’s seminal 2010 book Resonate focuses on visual storytelling. With the use of what she calls ‘sparklines’ she graphically illustrates the arc of a speech and maps the emotion and delivery by tracking the laughter and applause alongside the words (lyrics) of great speeches such as Martin Luther King’s “I have a dream” speech. It would be fascinating to apply a sparkline mapping audience reaction to live musical performances.

Political speechwriting

Political pundit and speechwriting professor Bob Lehrman (literally) wrote the book on speechwriting. In examining why persuasive speeches work, he lists the various forms of repetition (anaphora, epistrope, antimetabol, and climatic order) and forms of vivid language (simile and metaphor, understatement and irony) that work on the heartstrings, not just the head space, of the audience.

While these techniques can be employed by poets, they resonate differently when delivered by a compelling speaker.

In the beginning was the word

In claiming that “everything that happens in a poem has to happen within the text, in silence really” Armitage overlooks the pre-literate origins of poetry, when bards of old, the minstrels of Medieval times, recited epic poems which the tellers knew by heart. The literary techniques they used provided aide memoirs for themselves and for their listeners.

Writing in the New York Times, history professor Molly Worthen notes:

Since ancient times, humans have memorized and recited poetry. Before the invention of writing, the only way to possess a poem was to memorize it. Long after scrolls and folios supplemented our brains, court poets, priests and wandering bards recited poetry in order to entertain and connect with the divine.

Sorry, Simon, the origins of the poem are with the word, not the page.

Robin Wigglesworth highlights the effect of Artificial Intelligence on executive communications in an article in the Weekend FT (subscription required). A new form of “robo-surveillance” by trading algorithms is spurring executives to place a deeper focus on the spoken word.

Executives are coached to avoid saying certain phrases, such as “but” which could trigger stock sales by natural language processing (NLP) machines taught the intricacies of human speech. Using NLP, investment funds instantaneously scrape speeches, social media chatter and corporate earnings calls for clues.

Cat and mouse

The result is a cat and mouse game, where CEOs try to outwit the machines that can pick up a verbal clue that a human might not even realize is relevant.

A recent academic paper — How to Talk When a Machine Is Listening: Corporate Disclosure in the Age of AI — points out that companies are keen to show off their business in the best possible light.

…firms with high expected machine downloads manage textual sentiment and audio emotion in ways catered to machine and AI readers, such as by differentially avoiding words that are perceived as negative by computational algorithms as compared to those by human readers, and by exhibiting speech emotion favored by machine learning software processors.

The paper found that companies have tweaked the language of annual reports and how executives speak in public to avoid words that might trigger red flags for machines listening in.

These changes extend to the tone of voice executives use, in addition to the words they use. The paper notes:

Managers of firms with higher expected machine readership exhibit more positivity and excitement in their vocal tones, justifying the anecdotal evidence that managers increasingly seek professional coaching to improve their vocal performances along the quantifiable metrics.

Some companies’ investor relations departments are even running multiple draft versions of press releases and speeches through such algorithmic systems to see which scores the best.

In return, NLP powered algorithms are also continuously adjusted to reflect the increasing obfuscation of corporate executives, so it ends up being a never-ending game of fruitless linguistic acrobatics.

In this game, the machines have the upper hand. The algorithms can immediately adjust for a chief executive’s idiosyncratic styles.

A certain CEO might routinely use the word “challenging” and its absence would be more telling while one that never uses the word would be sending as powerful signal by doing so.

Body language

Machines are still unable to pick up non-verbal cues, such as a physical twitch ahead of an answer, but experts predict it’s only a matter of time before they can do this as well.

Where this will all end, and the impact it will have on speechwriters, presentation coaches, investor relations, and PR professionals is open to speculation.

Usually when machine and men collide, it’s the machines that have the upper hand.

Jeff Nussbaum and Kate Childs Graham, the 2020 Democratic convention speechwriters, have written a fascinating article in the Washington Post detailing how the ‘Zoom era’ has radically transformed political speechwriting.

While this probably won’t cause Bob Lehrman to tear up the guidance in his excellent book The Political Speechwriter’s Companion, it shows how the future of political rhetoric has been affected by the pandemic that required the prerecorded speakers at the convention to deliver speeches without a stage, an arena, or a live audience.

More is less

Nussbaum and Graham list the speechwriting techniques they used to script remarks for maximum impact, including:

• Speaking at 150-170 words per minute vs. the 125 typical when speakers in front of a live audience pause for laughter or applause.
• Cutting extraneous content to fit in tight 2 1/2 minute timeframes (the average length of a speech at this virtual convention).
• Dropping the rhetorical techniques of “call-and-response” or “litany” (eg. ending each section with a phrase like “Yes, we can.”)
• Delivering the headline message upfront, not burying the message in a lengthy speech.
• Dealing with the loss of the lectern — as TED talks have. Absent that visual crutch bestowing authority on the speaker, the venue supplemented the message: Kasich at the crossroads; Jill Biden in the classroom she once taught in.
• Invoking feelings via storytelling “As neuroscientist Antonio Damasio once put it, humans are feeling machines that think, not thinking machines that feel.” Hence the more memorable remarks were delivered by everyday people — the young man who stuttered, the lady whose father had believed Trump’s message on COVID-19 and died for his beliefs.

Michelle

Ironically, the one speech the professional writers did not script was the one many consider among the most powerful — delivered by Michelle Obama. The authors note:

She didn’t speak to 20 million television viewers: she spoke to one viewer in an intimate conversation that happened to take place 20 million times.

Speechwriting In Theory and Practice, by Jens E. Kjeldsen, Amos Kiewe, Marie Lund and Jette Barnholdt Hansen

Reviewed by Neil Hrab, Rhetoric Editor, Vital Speeches of the Day

A Book We’ve Long Awaited

Speechwriters have hoped for a long time to see a book like this appear in print. Speechwriting in Theory and Practice’s 13 chapters are grounded in a combination of academic perspectives on the evolution of rhetoric and persuasive speech, alongside a close study of how speechwriters and speakers collaborate, in the real world, to prepare remarks for delivery. In addition to the usual White House anecdotes, we are also treated to superb insights from the lived experience of European speechwriters.

Similar books have tried hard to strike the same balance, with well-intentioned, if somewhat uneven results. Clearly, the authors of this book have studied these earlier works, and charted a different course. Their joint introduction modestly proclaims on page 2 that “[o]urs is not a handbook [on speechwriting]. There are plenty of these.” Don’t take this declaration at face value, however – the book unites both theory and practice in such a way that any newcomers to the speechwriting field will find it of great practical value (particularly Chapters 6 to 13).

The great strength of Speechwriting is that it flips the typical approach for an academic study of speechwriting on its head. That more conventional structure might focus on topics such as how various rhetorical devices identified 2000+ years ago in ancient Greece and Rome continue to appear in 21st century speeches, or tout various still-relevant insights from Aristotle’s writings on rhetoric. (Aristotle makes periodic appearances throughout this book, more as a bystander, however, rather than as inescapable off-stage narrator.)

Speechwriting in Theory and Practice fully acknowledges this fascinating continuity in rhetorical practice over the centuries, but then takes the reasonable position that, since that continuity is already very clear—why not explore aspects of contemporary speechwriting that are less well known, at least outside of professional circles?

By opening its pages to consideration of contemporary speechwriting practice, the book raises questions pointing to promising areas for future research. These include: Why is corporate speechwriting so “astonishingly under-researched” compared to political speechwriting? (See Chapter 6) How does the need for a government speechwriter to get the approval of “higher levels in the system” shape the drafting of official speeches? (See Chapter 7.) Why, exactly, can PowerPoint be so deadly when it comes to trying to hold an audience’s attention? (See Chapter 10, especially the observation that the real problem posed by digital presentation tools like PowerPoint is that they invite “the speechwriter to present statements, not to reflect, make arguments, and tell stories.”) The discussion in of the potential ethical challenges faced by speechwriters in Chapter 11 is stimulating as well.

It’s no exaggeration to say that the authors of Speechwriting in Theory and Practice have blazed a new trail. With that path now open, let us hope others will follow in their steps, and open up this exciting new territory further.

This review first appeared in Vital Speeches of the Day and is reprinted here by express permission.

I caught a rebroadcast of an Intelligence Squared debate on NPR last night.

Intelligence Squared is a weekly forum for balanced and intelligent debate. Their goal is to restore critical thinking, facts, reason, and civility to American public discourse. (Hurray for them, although I can’t help thinking they’re pissing into the prevailing wind these days.)

The episode I tuned into was first broadcast February 11, 2019 and featured a unique debate between a world-class champion debater and an Artificial Intelligence (AI) system developed by IBM called Project Debater.

While the program was a fascinating glimpse into the rather arcane world of debating, it was the implications for speechwriters that had me wondering if we’re fast approaching the time when AI will replace the need for a human being to be involved in writing well constructed, persuasive speeches.

Project Debater

Project Debater faced off against Harish Natarajan, a grand finalist at the 2016 World Debating Championships and winner of the 2012 European Debating Championship. His opponent at the IBM Think conference in San Francisco was Project Debater — a two-metre-tall black box. She spoke in an American female voice through a blue, animated mouth. IBM claims that it’s the first AI system that can debate humans on complex topics.

Despite the rather uncanny resemblance to the black obelisk that features in the opening scene of Kubrick’s 2001 A Space Odyssey, the system exhibits none of the menace of the HAL system in that dystopian view of AI.

How it works

Project Debater is designed to debate humans on complex topics using a combination of pioneering research developed by IBM, including: data-driven speechwriting and delivery, listening comprehension, and modeling human dilemmas.

It goes beyond the Watson system that famously beat a human on the TV Quiz show Jeopardy, which showed a machine could respond to open-ended questions.

Project Debater digests massive texts, constructs a well-structured speech on a given topic, delivers it with clarity and purpose, and rebuts its opponent. Eventually, IBM claim, Project Debater will help people reason by providing compelling, evidence-based arguments and limiting the influence of emotion, bias, or ambiguity.

Isn’t that part of your job description, speechwriters?

Speechwriting by numbers?

The IBM website does not mince words about the capabilities of their system:

Project Debater relies on three pioneering capabilities. The first is data-driven speech writing and delivery, or the ability to automatically generate a whole speech, reminiscent of an opinion article, and deliver it persuasively. The second is listening comprehension, which is the ability to understand a long spontaneous speech made by the human opponent in order to construct a meaningful rebuttal. The third is the system’s ability to model human dilemmas and form principled arguments made by humans in different debates based on a unique knowledge graph. By combining these core capabilities, it can conduct a meaningful debate with human debaters.

Of note:

• Unlike speechwriters today, the AI system had no access to the internet.
• The massive amount of stored data it deployed was logically organized and presented in a fraction of the time most writers would take.
• Impressively, the system created an argument that clearly supported the side of the argument it was assigned.
• The system synthesized input from the debate opponent — a win when a speaker needs to respond, as in political dialog.
• Further, the AI system could easily frame both sides of a debate, helping speechwriters anticipate opposing points of view.

Overall, a sobering development.

Team player?

The question is, can speechwriters look to AI systems to be team players? Many would appreciate it if AI did the leg work assembling facts and blocking out the basic arguments. They could then give a final polish to the machines’ draft.

Or will an AI enabled speech writing entity get to the point where it will refuse to “open the pod bay door” and leave the human out in the cold? What job security can speechwriters expect in a world where AI systems can create persuasive, logical, and well-researched speeches in a fraction of the time a human could?

What say you?

On Thursday, December 13 the Silicon Valley Speeechwriters Roundtable hosted Ian Griffin in a free conference call.

I’m usually the one doing the interviewing. However, Barbara Seymour Giordano suggested she turn the tables and interview me. Barbara had been my guest back in May.

Barbara and I discussed my career as a freelance speechwriter, corporate employee, and blogger. We reviewed how I got into speechwriting, my experience in Silicon Valley companies, my Professionally Speaking blog as well as my new Booch News venture, and more.

To hear the full discussion, click on the podcast icon below.

Podcast: Play in new window | Download (Duration: 49:46 — 11.4MB)

Subscribe: Apple Podcasts | RSS

On Thursday, August 30th the Silicon Valley Speechwriters Roundtable hosted Jeff Davenport in a free conference call.

Jeff serves as an executive speaker coach and senior content developer at Duarte, the well-known communication design and consulting firm based in Santa Clara, founded by Nancy Duarte.

Using his background as a screenwriter and professional public speaker, Jeff helps clients communicate powerfully and persuasively by infusing story, dynamism, and empathy into their presentations. Whether he’s coaching high-level executives or thought leaders taking the stage for conference keynotes or commencement addresses, Jeff brings a thoughtful, personal touch to his roles, tapping into speakers’ personal passions and helping them create lasting connections with their audiences.

Jeff is a 2017 Cicero Award winner in the Public Policy category for his speech ‘Someday is Today’ delivered by Carlos Rodriguez-Pastor at CADE in Lima, Peru.

The call covered a wide range of topics including:

• How he went from a wallflower in high school to a premier public speaking coach.
• The secrets of the “Duarte Method” that any and all speechwriters can employ (Hint: read Resonate and Illuminate).
• The value of the DataStory training workshop available from Duarte that helps speechwriters structure a compelling argument based on analytical data.
• The three books on screenwriting he recommends speechwriters read:
1. Screenplay: The Foundations of Screenwriting, by Syd Field
2. Save the Cat!, by Blake Snyder
3. Into the Woods, by John Yorke
• What you’ll learn by watching the directors cut of Toy Story 3 on Blu-Ray.

Jeff’s parting words:

I would encourage anybody to do more public speaking, especially if you are writing for other people. We all had PE teachers who we realized never once played a sport. They were terrible PE teachers. So get out there and know what it’s like to play. Take a public speaking class. It doesn’t matter what it’s about. Do some sort of public speaking, writing for yourself and delivering yourself so you can get more in the heads of your clients and what know their true struggles are.

Otherwise, I would add, you’re forever the virgin trying to write a sex manual, aren’t you?

To hear the full discussion click on the podcast icon below.

Podcast: Play in new window | Download (Duration: 1:15:38 — 17.3MB)

Subscribe: Apple Podcasts | RSS

Two letters from Wednesday’s edition of the Financial Times attribute a quote about the difficult of writing a short speech or letter rather than a much longer one to two different sources.

Pascale’s quote is translated as “The present letter is a very long one, simply because I had no leisure to make it shorter.”

Woodrow Wilson’s quote is variously represented as

If I am to speak ten minutes, I need a week for preparation; if fifteen minutes, three days; if half an hour, two days; if an hour, I am ready now.

The debate over who to attribute the quote to has been analyzed in depth by the Quote Investigator — an invaluable resource — which finds evidence for a number of sources including Woodrow Wilson; Abraham Lincoln; Rufus Choate; Thomas B. Macaulay; William Howard Taft and Mark Twain.

No matter the degree of difficulty, the final word on the value of short speeches goes to British Prime Minister Winston Churchill:

A good speech should be like a woman’s skirt; long enough to cover the subject and short enough to create interest.

However, the Quote Investigator finds that Churchill was re-stating a saying that had been in circulation for over 20 years.