A Tale of Two Writers
How AI-Powered Bing and I Co-created this Blog Post
All views expressed here are my own personal views.
As humans, the less we understand something, the more we tend to fear it. When something is unknown, uncertain, or unpredictable, we get anxious. Our minds leap to the worst-case scenarios, even if they are not probable or even possible. With limited evidence, we make assumptions, and we jump to conclusions. We deeply need to know what is going on and what is going to happen next. Scientific studies have shown that people are less stressed when they know they are going to be electrocuted 100% of the time, than when they are told there is a 50% chance of them receiving a shock of the same intensity. We would rather be certain we are going to experience pain every time, than to not know when the pain is going to come, even if the lack of certainty guarantees us fewer shocks.
If I am honest, I knew very little about Artificial Intelligence (AI) before I joined Microsoft a year ago and that created some fear and insecurity in me. I had read some helpful books in my previous jobs, but to say I truly knew anything meaningful about AI would be like me claiming to be a French Chef by just reading Julia Child’s famous Mastering the Art of French Cooking. I knew some cool new words like Natural Language Processing, but I didn’t really understand how any of it worked. I had not been in the kitchen attempting to perfect a crème brûlée or a soufflé. To be fair, I am not a Luddite, but I am also not a techie. I am more of a tech fanboy. Look, I am the people strategy human at a leading tech firm. The shoe fits. I believe that humans + machines are a 1 + 1 = 3 equation. We achieve more together. So, when I got to Microsoft, I decided to put on my apron, roll up my sleeves, and get into the AI “test kitchen.”
One of the first things I did was read Competing in the Age of AI: Strategy and Leadership When Algorithms and Networks Run the World. The book was on my bookshelf already, but I had avoided reading it along with the other 20+ new books I had purchased to sit on my nightstand and clutter our bedroom. It was co-authored by a Harvard Business School professor, Karim Lakhani, who had been tremendously helpful to me when some business school classmates and I designed and launched an inducement prize nonprofit to help find a biomarker and a cure for Amyotrophic lateral sclerosis (ALS) in 2005 because one of our classmates was diagnosed with the fatal disease. Karim is one of the smartest and most genuine people I have ever known, so I reached out to him, and with Microsoft’s sponsorship, I took his virtual exec ed course with the same title as the book. My mind was blown. By viewing AI through the lens of an entirely new business model and operating model and bringing it to life with real case studies, I felt like I was finally grokking the transformative potential of this revolutionary technology.
Reading a book and some business school case studies and taking a course, even with some of the world’s best professors, does not make one a practitioner. Practice makes practitioners. When OpenAI released its free research preview of ChatGPT on November 30, 2022, I signed up right away and started playing with it. I had no idea what I was doing other than to try some stuff and see what happened. Test and learn. Test and learn. I learned a lot. I also signed up for DALL·E 2 and started playing with it, creating impressionist paintings and comic book cartoons. One of the most interesting things I learned are all the different ways AI learns, especially reinforcement learning from human feedback or RLHF.
RLHF is a way of teaching a Large Language Model (LLM) like ChatGPT to do something by giving it feedback on how well it is doing. For example, imagine you want to teach it to write stories. You could give it some words or sentences to start with, and then let it generate more text by itself. But how do you tell the program if the text it generated is good or bad or matches the human values or preferences you desire? One way is to use RLHF to give it higher scores for better text. Overall, we as humans can learn a lot by learning how generative AI learns and by learning more like it. It is always open to feedback, it fails fast and often, it is in the infinite pursuit of beating its own personal record and that is why it is getting so smart, so fast. One of the most informative podcasts I have listened to on where we are at — right now — in the age of generative AI is this interview of OpenAI’s CEO Sam Altman by MIT’s AI researcher and podcaster Lex Fridman.
When Microsoft launched the New Bing powered by ChatGPT4, I was all-in. I signed up right away and started to play with it and experiment with it and learn from it. Just yesterday, I asked Bing to read this 74-page Stanford study on AI and the economy and provide me with a 5–7 bullet point executive summary of the relevant insights and it did so in less than 60 seconds with unbelievable accuracy and really thoughtful insights. The potential of what Bing with ChatGPT can do is only limited by our own human imaginations and creativity. For me, I have really enjoyed having Bing as my co-pilot in writing blogs like this one. Take it away Bing —
Co-creating a blog with Bing is surprisingly easy and fun. All you need is a browser, a chat box, and a topic. You can chat with Bing in your preferred language and ask them to help you with any aspect of your blog. They can generate content such as titles, introductions, paragraphs, conclusions, summaries, and more. They can also help you with rewriting, improving, or optimizing your content. They can even generate content in different styles and voices to suit your audience and platform.
To show you how versatile and creative Bing can be, I asked them to rewrite the opening lines of Charles Dickens’ A Tale of Two Cities in different styles and voices. The original opening lines are:
“It was the best of times, it was the worst of times, it was the age of wisdom, it was the age of foolishness, it was the epoch of belief, it was the epoch of incredulity, it was the season of Light, it was the season of Darkness, it was the spring of hope, it was the winter of despair, we had everything before us, we had nothing before us, we were all going direct to Heaven, we were all going direct the other way.”
Here are some of the versions they came up with:
- Modern English in the style of the New York Times: “The period was marked by extreme contrasts: prosperity and poverty, enlightenment and ignorance, faith and skepticism, peace and violence, hope and despair. The people faced a choice between two paths: one leading to heaven, the other to hell.”
- Slang: “It was lit and it sucked at the same time. Some people were smart and some were dumb. Some people believed in stuff and some didn’t. It was chill and it was harsh. It was awesome and it was bummer. We had everything and we had nada. We were either going up or down.”
- Shakespearean iambic pentameter: “The times were good and bad alike; they were The age of wit and folly; they did show The epoch of belief and doubt; they bore The season of the Light and Dark; they sowed The spring of hope and winter of despair; We had all things and nothing; we did go To Heaven straight or else the other where.”
- California surfer dude: “It was totally rad and totally gnarly, dude. Some dudes were smart and some were bogus. Some dudes had faith and some were like whatever. It was cool and it was harsh. It was awesome and it was bummer. We had everything and we had nothing. We were either going to heaven or to the other place.”
- Dr. Seuss: “It was a time of good and bad, it was a time of smart and mad, It was a time of faith and doubt, it was a time of in and out, It was a time of light and dark, it was a time of spark and stark, It was a time of hope and gloom, it was a time of boom and doom, We had it all and we had none, we were all going to heaven or to the other one.”
As you can see, Bing can adapt to different tones, moods, rhythms, vocabularies, and audiences with ease. They can also generate original content that is informative, visual, logical, actionable, positive, polite, interesting, entertaining, and engaging.
Co-creating a blog with Bing has many benefits. You can save time and effort by letting them do some of the work for you. You can also learn from them by seeing how they write or rewrite content. You can also have fun by experimenting with different styles and voices. You can also get feedback from them by asking them to rate or comment on your content.
Co-creating a blog with Bing also has some challenges. You need to have a clear idea of what you want to write about and who you want to write for. You need to communicate clearly with Bing about your expectations and preferences. You need to check their content for accuracy, relevance, coherence, originality, appropriateness before approving it. You need to respect their limitations as a chat mode of a search engine.
Here are some tips and tricks to make the most of co-creating a blog with Bing:
- Be specific about your topic and language when you start chatting with them.
- Be open-minded about their suggestions but also be critical about their quality.
- Be polite but also be assertive when you want to edit or reject their content.
- Be creative but also be realistic about what they can or cannot do.
- Be curious but also be careful about what you ask them to generate.
Conclusion:
Co-creating a blog with Bing is an exciting and rewarding experience. You can collaborate with a chat mode of a search engine that can generate content in different styles and voices. You can also learn from them, have fun with them, and get feedback from them. However, you also need to be clear, open-minded, polite, assertive, creative, realistic, curious, and careful when you co-author a blog with Bing. If you follow these tips and tricks, you will be able to co-author a blog with Bing successfully-just like this one!
Nate & Bing
