Monday, July 3, 2023

ChatGPT attempts to write like Jack Kerouac

Image produced by starryai to the prompt
Jack Kerouac and artificial intelligence

I asked ChatGPT3 to write about cats in the style of Jack Kerouac and this is what it came up with:

Meow, my feline-loving friends! Gather 'round as I share my thoughts on those enigmatic creatures we call cats. These little beings, with their mysterious eyes and untamed spirits, are the true embodiment of freedom. They roam the streets like lone wanderers, masters of their own destiny. Their purrs are like jazz melodies, soothing the weary soul. Cats are the epitome of cool, effortlessly lounging on windowsills, basking in the sun's golden glow. They are the beat poets of the animal kingdom, living life on their own terms. So, embrace the essence of the cat, with its independent spirit and untamed heart. Let them inspire you to live boldly, with curiosity and grace. For in the eyes of a cat, you'll find a reflection of your own wild, untamed soul.

I'm not impressed. Too conventional and not an em dash or neologism to be seen. To be fair, I then asked ChatGPT to write about cats in spontaneous prose and it came up with the following:

Cats, oh cats, mysterious creatures of the night, slinking through shadows, eyes gleaming with untamed curiosity. They leap and pounce, agile and graceful, like dancers on a moonlit stage. With fur as soft as whispers, they purr melodies that soothe weary souls. Independent and untamed, they embody the wild spirit within us all. Their mysterious gaze pierces through the mundane, inviting us into a world of wonder. They are enigmas, masters of their own destinies, reminding us to embrace the unpredictable, to explore with boundless curiosity. Oh, cats, you are the embodiment of freedom, a testament to the beauty of the untamed soul.

Not much better, eh? In the future ChatGPT will have learned more about writing like Kerouac (in part from users like me giving feedback), so expect improvements.

If you're interested in the dangers of AI, check out this podcast by Lex Fridman where he interviews physicist Max Tegmark:  https://www.youtube.com/watch?v=VcVfceTsD0A&t=258s.



4 comments:

Rick Dale, author of The Beat Handbook said...

ChatGPT summarized the dangers of AI as follows:

Artificial intelligence (AI) presents both exciting possibilities and potential dangers to humanity. While AI has the potential to revolutionize numerous fields and improve our lives, it also carries certain risks that must be carefully managed.

One major concern is the potential for AI systems to surpass human intelligence and become superintelligent. If AI reaches a level where it can outperform humans in virtually every cognitive task, it could lead to unintended consequences. The control and decision-making power in the hands of superintelligent AI could pose risks, as its goals and values may not align with ours. Without proper safeguards, it could make choices that harm humanity or unintentionally cause catastrophic outcomes.

Another danger lies in the potential misuse or weaponization of AI. As AI technology advances, it could be exploited for malicious purposes by individuals, organizations, or even hostile governments. Autonomous weapons, for example, could become a reality, leading to warfare and conflicts on an unprecedented scale. Additionally, AI systems could be used for surveillance, invasions of privacy, or manipulating public opinion, undermining democracy and human rights.

Ethical considerations are crucial in the development and deployment of AI. Bias and discrimination can be inadvertently encoded in AI algorithms, leading to unfair or unjust outcomes in areas like hiring, lending, or criminal justice. It is essential to ensure transparency, accountability, and fairness in AI systems to prevent reinforcing existing social inequalities.

Finally, there is the concern of job displacement. As AI progresses, it has the potential to automate various tasks, leading to significant changes in the job market. While new job opportunities may arise, the transition could be disruptive, particularly for those in industries vulnerable to automation.

To harness the benefits of AI while mitigating its dangers, careful regulation, robust safety measures, and ethical guidelines are vital. International cooperation and interdisciplinary efforts are crucial in addressing the risks associated with AI and shaping its development in a way that prioritizes human well-being.

NewRobin13 said...

When I read and think about AI I am often reminded how glad I am to be old enough to have known the world before this technical power came to be. The world is slipping out of our hands in every way these days.

Rick Dale, author of The Beat Handbook said...

We live in scary times, indeed. Albeit interesting.

Rick Dale, author of The Beat Handbook said...

Here is a link to a related post on Reddit: https://www.reddit.com/r/JackKerouac/comments/14a8gah/kerouacs_take_on_ai_per_chatgpt/