People worry that AI will be the death knell of creativity. I think AI will bolster creativity by bringing in new creators into the market and by expanding the boundary of creative experimentation.
I agree that creativity will be enhanced. The flavor might be different, but artist will find a way to art, and nothing gets them going like a new canvas to play with.
Thanks for the insights, Kartik. "It was better fifty years ago" is what people of my generation would say. A similar age group told the same fifty years ago. We are scared of changes. The side effects of inventions considered the most dangerous have changed the world for the better. Explosives, nuclear bomb, weapon systems technology during wars ...
Until the technology matures and stabilizes, there would be some problems and much bickering, including mine. In a test, about 100 students copied solutions from chatGPT and could not even type out the solutions in their answer books. As they copied from common sources, plagiarism counts were above 80% across their solutions. The teacher learned a lot but the students continue to crib that they have been judged unfairly with low scores while the answers could be correct. A bigger mess.
The culture we build and our commitment and professionalism make a difference.
Campus Fund Jun 21 - Dec 22 showed that BITS alumni have founded maximum companies, far ahead of IITs, IIMs, NITs and other big name institutions. RVU professors had some comments on how BITS gives absolute freedom to the students making the students more responsible. In other institutions with 40% of the seats are based on cast-cum-merit and the rest sold at exorbitant cost (management quota) creates a mess of culture. I have met students buying degrees so as to demand higher dowries. They would also buy marks by means unfair. chatGPT is an easy way out to avoid learning but fulfil the obligations on paper.
Any tool if put to unfair use can damage. We cannot stop building new tools with that fear.
Your point about students using these tools to complete tasks but ending up with a poorer understanding of those concepts is interesting. The Github Copilot study that I mentioned in post #1 shows something similar. Copilot increased developer productivity but developers had a poorer understanding of their code.
As informative and interesting as the earlier newsletter, Kartik. Your break down of what generative AI can mean for creative pursuits really helps to get into that frame of thinking. Clearly there are significant benefits as creative tasks become "easier" to achieve. I do believe change (in this case brought on by AI) is inevitable, necessary too especially if it is potentially an opportunity equalizer.
That said I can't help wondering.... the MIT study focused on creative writing tasks that are at a somewhat basic level when it comes to 'being creative', for its quality. The Microsoft study is more specific to coding. Are there any studies that go deeper into the 'finer aspects' of creativity and how they are impacted? to understand for example:
- is there a difference between the creator's "own" (for want of a better word) creativity and "borrowed" creativity? Does it matter at all? Or perhaps there are situations where this does/should matter?
- is/will there be a trade off between efficiency and imagination?
- the nuance, the rich depths, the romance in creating a song, a painting or writing a book.... ... will these be affected?
Wow, I am blown away by these comments. I think these are the right set of questions and I don't have good answers. At least, I have not seen any good in-depth study on any of these questions.
For me, a casual writer, I find the efficiency gains in drafting really frees me to spend more time imagining. It allows me to spend more time on the fun parts of creativity (ideating, digging deeper into the idea, etc) as opposed to being overwhelmed by the actual task of drafting, etc. So I am hoping there doesn't have to be a tradeoff between efficiency and true creativity.
I hope some day to have better answers but for now you have given me a great set of issues to think about.
I'm thrilled my rambling made sense to you! I do fully see how in many situations AI can aid the level and extent of creativity and indeed free the mind up for more critical creative tasks. And also to be honest but for reading your newsletter I wouldn't have thought about these issues at all. Thanks for that and looking forward to reading more from you!
Thank you for your posts, and for your book on machine intelligence which I as a novice learned many new things from. I subscribe to your idea that creative possibilities will not necessarily be diminished, and have faith in human ingenuity to use tools to push the needle forward. Yet, I'm curious what your views are about the "consuming public" or those that appreciate creative endeavors. Part of the enjoyment of a screenplay, a piece of music, a nicely written essay, or even poetry comes from knowing that someone likely spent years crafting their skill in order to perform at a level worthy of notice by others. In other words, will our knowing that efficient machine learning was responsible for something breathtaking eventually erode the standards of what we might consider creative, interesting, beautiful (usually the outcome of long hours of human labor, toiling, and focus in a single or set of manageable directions as opposed to an all expansive kaleidoscope made possible by generative AI).
Thank you for writing, eagerly waiting to see subsequent posts.
Great point, Sudev. I agree that we often value creative content because of the story behind the creation itself. In fact, I have seen at least two recent studies that show something similar. My colleague Stefano Puntoni says this in a recent WSJ post based on his research: "companies should also consider demand-side factors, such as the preference for human labor... This way, they can identify the business case for human employment, even when supply-side factors may speak against it...The key may be companies’ ability to let consumers know when products are made by humans—from packaging or advertising, for instance, or in-store displays." https://www.wsj.com/articles/ai-products-marketing-human-made-14a1f7e1?st=ii2kilng8aieqiz&reflink=article_copyURL_share
My next post will touch upon why I believe high-end human creativity will continue to thrive. More on that later this week.
+100
I've found a couple of times it's even helped me brainstorm and think more clearly.
I agree that creativity will be enhanced. The flavor might be different, but artist will find a way to art, and nothing gets them going like a new canvas to play with.
Thanks for the insights, Kartik. "It was better fifty years ago" is what people of my generation would say. A similar age group told the same fifty years ago. We are scared of changes. The side effects of inventions considered the most dangerous have changed the world for the better. Explosives, nuclear bomb, weapon systems technology during wars ...
Until the technology matures and stabilizes, there would be some problems and much bickering, including mine. In a test, about 100 students copied solutions from chatGPT and could not even type out the solutions in their answer books. As they copied from common sources, plagiarism counts were above 80% across their solutions. The teacher learned a lot but the students continue to crib that they have been judged unfairly with low scores while the answers could be correct. A bigger mess.
The culture we build and our commitment and professionalism make a difference.
Campus Fund Jun 21 - Dec 22 showed that BITS alumni have founded maximum companies, far ahead of IITs, IIMs, NITs and other big name institutions. RVU professors had some comments on how BITS gives absolute freedom to the students making the students more responsible. In other institutions with 40% of the seats are based on cast-cum-merit and the rest sold at exorbitant cost (management quota) creates a mess of culture. I have met students buying degrees so as to demand higher dowries. They would also buy marks by means unfair. chatGPT is an easy way out to avoid learning but fulfil the obligations on paper.
Any tool if put to unfair use can damage. We cannot stop building new tools with that fear.
Your point about students using these tools to complete tasks but ending up with a poorer understanding of those concepts is interesting. The Github Copilot study that I mentioned in post #1 shows something similar. Copilot increased developer productivity but developers had a poorer understanding of their code.
As informative and interesting as the earlier newsletter, Kartik. Your break down of what generative AI can mean for creative pursuits really helps to get into that frame of thinking. Clearly there are significant benefits as creative tasks become "easier" to achieve. I do believe change (in this case brought on by AI) is inevitable, necessary too especially if it is potentially an opportunity equalizer.
That said I can't help wondering.... the MIT study focused on creative writing tasks that are at a somewhat basic level when it comes to 'being creative', for its quality. The Microsoft study is more specific to coding. Are there any studies that go deeper into the 'finer aspects' of creativity and how they are impacted? to understand for example:
- is there a difference between the creator's "own" (for want of a better word) creativity and "borrowed" creativity? Does it matter at all? Or perhaps there are situations where this does/should matter?
- is/will there be a trade off between efficiency and imagination?
- the nuance, the rich depths, the romance in creating a song, a painting or writing a book.... ... will these be affected?
Wow, I am blown away by these comments. I think these are the right set of questions and I don't have good answers. At least, I have not seen any good in-depth study on any of these questions.
For me, a casual writer, I find the efficiency gains in drafting really frees me to spend more time imagining. It allows me to spend more time on the fun parts of creativity (ideating, digging deeper into the idea, etc) as opposed to being overwhelmed by the actual task of drafting, etc. So I am hoping there doesn't have to be a tradeoff between efficiency and true creativity.
I hope some day to have better answers but for now you have given me a great set of issues to think about.
I'm thrilled my rambling made sense to you! I do fully see how in many situations AI can aid the level and extent of creativity and indeed free the mind up for more critical creative tasks. And also to be honest but for reading your newsletter I wouldn't have thought about these issues at all. Thanks for that and looking forward to reading more from you!
Dear Kartik,
Thank you for your posts, and for your book on machine intelligence which I as a novice learned many new things from. I subscribe to your idea that creative possibilities will not necessarily be diminished, and have faith in human ingenuity to use tools to push the needle forward. Yet, I'm curious what your views are about the "consuming public" or those that appreciate creative endeavors. Part of the enjoyment of a screenplay, a piece of music, a nicely written essay, or even poetry comes from knowing that someone likely spent years crafting their skill in order to perform at a level worthy of notice by others. In other words, will our knowing that efficient machine learning was responsible for something breathtaking eventually erode the standards of what we might consider creative, interesting, beautiful (usually the outcome of long hours of human labor, toiling, and focus in a single or set of manageable directions as opposed to an all expansive kaleidoscope made possible by generative AI).
Thank you for writing, eagerly waiting to see subsequent posts.
Sudev
Great point, Sudev. I agree that we often value creative content because of the story behind the creation itself. In fact, I have seen at least two recent studies that show something similar. My colleague Stefano Puntoni says this in a recent WSJ post based on his research: "companies should also consider demand-side factors, such as the preference for human labor... This way, they can identify the business case for human employment, even when supply-side factors may speak against it...The key may be companies’ ability to let consumers know when products are made by humans—from packaging or advertising, for instance, or in-store displays." https://www.wsj.com/articles/ai-products-marketing-human-made-14a1f7e1?st=ii2kilng8aieqiz&reflink=article_copyURL_share
My next post will touch upon why I believe high-end human creativity will continue to thrive. More on that later this week.