ChatGPT-The new paradigm shift

Network Science March 01, 2023

The phenomena!
2022 was a watershed year in evolving the way people interacted with Artificial Intelligence. DALLE-2 captured the imaginations of everyone in the world at a scale which would have made its namesake painter proud. ChatGPT wrote movie scripts, party planning ideas, code and explained millions of concepts to people without having to read 10 websites for understanding. In 4 months, the world found alternatives for Google search, photoshop, Stack Overflow, and script writing. I personally used some of these tools to get inspirations for pitch decks and feature sets for enterprise technologies and occasioned an interesting conversation between Shakespeare and Douglas Adams in the style of their characters.
This excitement goes beyond acceptance of new paradigms of search and creation. Proof of its popularity is evidenced by its incredible rate of adoption which till last count was about 100 Million monthly active users just on ChatGPT. The insanity compounds with the enormous flow of investments from Microsoft in Open AI with an aim to disrupt Google search. To keep pace with the new paradigm, Google has backed Anthropic as its own horse in the race.

Source:https://www.statista.com/chart/29174/time-to-one-million-users/
Shifting paradigms
Though the introduction of Artificial Intelligence on a global stage is not new, it feels a little different from what the past has shown us. In the age of AR/VR, we saw companies focusing bringing AR and VR to the forefront. Similarly, with blockchain and drones, the technology was at the center of the value proposition. However, AI/ML was playing the old sage, hiding its hard work behind an outcome that many believed could have been done with some basic coding and hard work themselves. In engaging with a website chatbot, or outputs from a predictive engine, users often viewed the output of complex analysis as something that can be created with enough time on excel or by searching hard enough. However, moving to generic text and image has changed that. AI’s abilities seem closer, and at times beyond, finished products of professionals. Search is not about narrowing your universe anymore, but finding exactly what you are looking for. Likewise, creation is now focused on bringing fantastical ideas come to life where the limiter is your ideas. This paradigm shift is powerful beyond what we can comprehend today. What seems like joyous tools to complete your job swiftly can open paths of infinite possibilities leading to democratised creation.

The Tomorrow
Today tools such as ChatGPT or Chatsonic are giving you the scaffolding around complete solutions which can be easily configured or edited to your designed results. However, getting more contextual results and providing the ability to ad-hoc dynamically will move the usage of these tools from starting points to being complete solution suits. Social media is already flooded with bundled AI tools for advertisers, marketers, designers and more. In the coming weeks, we will see more specialised uses of AI bundling. This will lead to a complex network of AI’s talking to each other possibly through a single interface. I know this might sound a bit like steps towards Terminator coming back to kill Sarah Conner, but till that happens, this is good news for creation. Such paradigms of AI can expand the scope of who we call creators. Lack of specialised software skills will not equate to a lack of quality. Tomorrow a diligent college student might be able to win an Oscar, or a grandmother set up a sustainable D2C brand. I obviously wax lyrical to a fanciful future, but stranger things have happened.
Even if I don’t think robot from the future is likely, it doesn’t mean there are not concerns regarding this new paradigm. This form of AI will limit growth of millions of medium to even high skilled jobs. There is a growing notion that many jobs will be replaced by skilled people who know how to use AI rather than AI’s on their own. There is truth to the notion, however one does question the value you would assign to expert work. There are other issues such as biased training, questions around ownership, ethical dilemmas, lack of transparency, and privacy concerns, but each of these points deserves an in-depth study on it’s own.

The Conclusion
Artificial Intelligence is here to stay and its adoption will have good and bad consequences. However, arming yourself with the right skills to harness the potential seems like the best course of action, at least in the near term. AI is already disrupting my job by getting me access to specialised information for which I would have to spend a couple of hours studying. However, using it without adding my experience and knowledge will also lead to my downfall. After all, DALL-E 2 can make beautiful paintings in the style of Eduard Munch, but, at least today, it can’t convey the raw inner turmoil behind The Scream. That still needs some ‘human’ editing.

The Scream by Edvard Munch

ChatSonic prompt: Make a painting depicting anxiety and inner turmoil in the style of Edvard Munch

This blog has been contributed by Nikhil Mahen, Deep Tech Consultant at Network Science. You can write to Nikhil at nikhil@networkscience.ai for any queries regarding this blog.