Artificial Intelligences (AIs) aren’t exactly a new concept: they have been around practically since the dawn of computer science, and instrumental to its development as well as an object of research in itself. They’ve also been haunting the geeky recesses—and a few others—of popular culture for quite some time now (hello HAL9000).
The recent rise of generative AI (GenAI), however, has done nothing less than revolutionise the field: DALL-E, introduced in January 2021, was the first widely-publicised generative AI, allowing anyone to create digital images through everyday language descriptions.
Midjourney followed in July 2022, based on the same text-to-image process. Thus was the world blessed with disturbingly realistic pictures of Pope Francis setting new fashion trends Balenciaga-style, or of Donald Trump running for his life in the streets of an alternate Washington, chased by police officers.
Then, in November 2022 ChatGPT popped up, taking generative AI to a whole new level—and with that, possibilities and business opportunities that we’re only starting to fathom. This conversational agent, or chatbot, trained with hundreds of gigabytes of books, articles, websites and other text available on the internet, can tap into this vast corpus of data to answer whatever request is submitted—well, almost.
To the greatest delight of students and consultants alike, it’s now possible to generate pages of essays and reports in a blink of an eye. Or to code computer programmes. Or to compose music. Or to write an article—by the way, let it be known that this blog entry has been written by human beings. Mostly.
We thought that this opportunity to rely on what’s basically a direct line to the entirety of human wisdom was too good to pass up, and that we could put ChatGPT to work on some of the metaphysical questions that keep exhausting the brightest brains: Am I loved? What is the best 90’s sitcom? And more pressingly: What on earth am I going to read this summer?*
After some brainstorming, we decided to focus on the latter, the most difficult question after all, especially with summer vacations looming. For good measure, we thought it would be interesting to call on our in-house AI specialists to share their own suggestions, and see where we stand compared to what the robot gave us. Here we go:
*For the curious cats among you, dear readers, the answers to the two first are “yes” and “Friends”.
What are you going to read while on holidays? Any suggestions for our readers?
Here’s what Saharnaz Dilmaghani, Senior Associate, Data Analytics and AI, at PwC Luxembourg told us:
“We’re living in exciting times! The way we think about AI has totally changed and we’re still just scratching the surface of what generative AI can do. This summer, I’m planning to read ‘Impromptu: Amplifying Our Humanity Through AI.’
And guess what? It’s co-authored by GPT-4! This book, penned by Reid Hoffman and our AI buddy GPT-4, explores how AI is changing the way we live, work, and think. It covers the good, the bad, and the downright funny, with discussions on human nature, original sci-fi plots, and more.
As for Reid Hoffman, the co-founder of LinkedIn and a former board member of OpenAI, his influential role in building several of today’s leading consumer technology businesses makes his insights particularly noteworthy.
I also plan to read ‘The Age of AI and Our Human Future’. This book is by a trio of big thinkers in policy, industry, and academia, and it’s all about how AI is changing society. It asks readers to think about the relationship between humans and AI and what it means for us all.
Happy summer reading!”
Here’s what ChatGPT replied:
“As an AI language model, I don’t have personal preferences or the ability to experience reading books. However, if I were to suggest a popular and widely acclaimed book, here’s a good recommendation:
‘The Immortal Life of Henrietta Lacks’ by Rebecca Skloot.
This book tells the remarkable and true story of Henrietta Lacks, an African American woman whose cells were taken without her knowledge or consent in 1951. These cells, known as HeLa cells, became one of the most important tools in medical research, contributing to numerous scientific breakthroughs. However, Henrietta Lacks and her family were largely unaware of her cells’ significance for many years.
‘The Immortal Life of Henrietta Lacks’ delves into the ethical, scientific, and personal implications of this case. It explores topics such as medical ethics, race, and the intersection of scientific advancement and human rights. The book offers a compelling narrative that blends science, history, and personal storytelling, shedding light on important ethical considerations within the medical field.
This non-fiction work has received critical acclaim for its engaging storytelling and thought-provoking examination of the impact of scientific discovery on individuals and society as a whole.
Remember to explore reviews, synopses, and personal interests to ensure the book aligns with your specific preferences. Enjoy your summer reading!”
Well, thank you for your suggestions Saharnaz and ChatGPT. They seem very interesting!
What read would you suggest for someone who’d like to learn more on the topic of generative AI specifically?
Here’s what Saharnaz suggested:
“Generative AI is turning heads these days, reminding me of when all students had to learn a bit of coding at university regardless of their major. Nowadays, knowing about generative AI is becoming just as important, and trust me, you don’t want to be left behind.
First up, let’s explore Prompt Engineering. It’s like finding a new way to communicate with AI. I came across a great book, ‘Prompt Engineering: Unlocking Generative AI: Ethical Creative AI for All’ by Navveen Balani. It’s all about mixing our creativity with AI to make cool things. If you’re curious about generative AI and want a fun learning project for the summer, this book is a good start.
Next, let’s talk about how Generative AI is impacting productivity. We all agree it’s helping, but how much, exactly? I found an interesting study by MIT, ‘Experimental Evidence on the Productivity Effects of Generative Artificial Intelligence’. They explored how much a tool like ChatGPT can boost productivity and how it impacts job satisfaction and self-confidence. It’s pretty cool to think about how this tech is changing the way we work and create.
Of course, with new tech comes new challenges. If you want to dig deeper, there’s a detailed study by Stanford, ‘On the Opportunities and Risks of Foundation Models’. It gives a balanced view of the potential, but also the challenges of these models. Definitely a must-read for a broader understanding of generative AI.
Just remember, all these are just the tip of the iceberg. The world of generative AI is like a runaway train, with new discoveries popping up everywhere. It blows my mind to think that all this research and these impacts happened in less than seven months, and I can’t wait to see where it goes.”
Our buddy ChatGPT also had something to say:
“If you’re interested in learning more about generative AI specifically, here are a few book recommendations that can provide you with valuable insights and in-depth knowledge on the subject:
- ‘Deep Learning’ by Ian Goodfellow, Yoshua Bengio, and Aaron Courville: This book covers various aspects of deep learning, including generative models, and provides a comprehensive introduction to the field.
- ‘Generative Deep Learning’ by David Foster: This book focuses specifically on generative models and deep learning techniques, providing practical examples and code implementations.
- ‘Generative Adversarial Networks Cookbook’ by Josh Kalin: This book offers hands-on guidance and practical recipes for building and training generative adversarial networks (GANs), a popular class of generative models.
- ‘Hands-On Generative Adversarial Networks with PyTorch’ by John Hany: This book focuses specifically on GANs and provides practical examples and implementations using the PyTorch framework.
These books offer different levels of technical depth and practical guidance, allowing you to explore generative AI from both theoretical and practical perspectives. Remember to check the prerequisites and choose the book that aligns with your current knowledge and learning goals.”
What developments can you foresee, if any, in the field of generative AI, for the next two to three months? (AKA “What am I going to miss while lying on the beach?”)
Now it’s Andreas Braun’s turn, who’s a Director, Data Analytics and AI, at PwC Luxembourg, to share his thoughts:
“Two months in AI-time feels like a decade these days. Nevertheless, even data scientists and AI researchers take summer breaks—shorter in the US, maybe. However, I can think of two AI-related tools to try out, and one to look forward to.
First, Midjourney. It’s currently the most powerful AI image generator, whose newly-released updated version can create ever more realistic photos. The picture below is an example of a Midjourney-created image, illustrating a beautiful—yet imaginary—summer hiking getaway. You can try it and flex your creative side for a monthly fee that just amounts to a glass of wine in Luxembourg restaurants.
The second tool that’s really worth a try is the newly-released OpenAI Code Interpreter. This tool (for paying customers only) enables you to become a software developer, or a data scientist, without any programming knowledge. You can have the AI generate charts from data sets, design your website or develop the next killer app. Your imagination is the limit! Here is an example of analysis by Code Interpreter, based on an Excel file with UFO sightings I uploaded.
What I’m looking forward to when returning after summer is the Office Copilot. Let’s be real: most of us spend way too much time dabbling in Word, Excel and Powerpoint, and Office Copilot’s promise is to remedy just that. I’m really curious to see how such integrated AI assistants might simplify my life…or make it more complex.”
If you are wondering what ChatGPT has to say about its own development, fear not. It answered:
“As an AI language model, I don’t have access to real-time information or the ability to predict specific future events. My training only goes up until September 2021, so I can’t provide insights on developments that have occurred after that date.
However, the field of generative AI has been advancing rapidly in recent years, and it is likely to continue evolving in the coming months. Some potential areas of development could include:
- Enhanced generative models: Researchers may develop new architectures or techniques to improve the performance and capabilities of generative models, leading to more realistic and diverse outputs.
- Domain-specific applications: Generative AI may find applications in specific domains such as art, fashion, music, or video games, enabling the creation of unique and creative content.
- Ethical considerations: As generative AI becomes more prevalent, there will likely be discussions and advancements in addressing ethical concerns, such as privacy, bias, and misuse of AI-generated content.
- Integration with other technologies: Generative AI may be combined with other technologies, such as augmented reality (AR) or virtual reality (VR), to create immersive and interactive experiences.
While these are potential areas of development, it’s important to note that the field of AI is highly dynamic, and breakthroughs can occur unexpectedly. To stay updated on the latest developments in generative AI and other AI-related fields, you may consider following research publications, attending conferences, or referring to reputable sources that cover AI and technology news. Enjoy your time on the beach, and if you have any more questions, feel free to ask!”
Conclusion
Well, that gives us plenty of food for thought—one could almost wish for MORE vacation just to read all these fine books. Many thanks to Andreas and Saharnaz for their suggestions and for playing the game of questions, and to ChatGPT for its patience and goodwill—it’s just a robot, but a very articulate one.
So, ChatGPT, would you like to say good-bye to our readership, and wish them a great summer?
“I’m sorry, I’m afraid I can’t do that.”
What we think
If there is one thing AI can’t do for you, it’s reading and learning. Well, it can help you by summarising and providing you with information in an easily digestible way, but in the end, what makes sense out of all the AI work is the human brain. So, AI will not take away your job anytime soon, but it might well change it, or even improve it, in meaningful ways. Be AI curious this summer, read up on the topic and try out the available tools. You may find some inspiration for your job and personal development… and end up ahead of the curve.
Andreas Braun, Director, Data Analytics and AI, at PwC Luxembourg