Are Knowledge Scientists Nonetheless Wanted within the Age of Generative AI?

Are Data Scientists Still Needed in the Age of Generative AI?
Picture Created with Steady Diffusion


Precisely 2 years in the past, I wrote an opinion piece referred to as “Knowledge Scientists Will likely be Extinct in 10 Years”. To my shock, it will develop into one in every of my most-read articles on each Medium and  KDnuggets. Nevertheless, the response was polarizing. It attracted probably the most criticism that I’ve obtained in my grownup life. I foretold the demise of the sexiest (and one of the vital in-demand) jobs of the twenty first century and my friends took challenge, however I accepted the suggestions and life moved on. Quick ahead to now; and what a distinction two years makes. ChatGPT has taken the world by storm, and with it, the narrative {that a} particular position shall be phased out has been eclipsed by one other with far higher implications; the obsolescence of human capital in each conceivable business. 

The revolution appears to have occurred in a single day. However these of us which have adopted the progress of deep studying intently know very properly that it didn’t. ChatGPT was the buildup of a long time of analysis that inexplicably culminated in an unassuming chatbot. On the core of the success of ChatGPT is the truth that it democratizes AI. Being code literate and having deep technical data are longer obstacles to entry, accessibility to cutting-edge deep studying has transcended the area of educational analysis and large tech to be obtainable on the fingertips of anybody with wifi entry and an e-mail deal with.



By no means in my wildest desires did I feel that we had been on the precipice of a technological revolution of the pace, scale, and nature that we skilled? Earlier than LLMs and Textual content to Picture Fashions, Generative AI (GAI) was largely synonymous with Ian Goodfellow’s Generative Adversarial Networks (GANs). It was hailed as one of many nice AI analysis contributions in recent times, manifesting within the capability to make use of a pair of neural networks to generate artificial, photo-realistic pictures. These of us which have labored with GANs earlier than know that they’re notoriously tough to coach and even when applied accurately, the use circumstances on the time had been restricted. Due to this fact, it’s much more wonderful that generative deep studying has heralded the most recent spherical of developments. 

So why would ChatGPT(and its GAI compatriots) convey information scientists to the brink of extinction? Let’s revisit the unique thesis from two years in the past:

  1. The flexibility to regurgitate code and use software program packages will now not outline a knowledge scientist as low/no-code options had been already turning into prevalent.
  2. The flexibility to work and analyze information will develop into an assumed talent set for a lot of roles very similar to computing abilities and MS Workplace data.
  3. On this paradigm area specialists that may remedy real-world issues will excel. Knowledge science will develop into a part of their toolkit.
  4. Given the above, generalist information scientists shall be phased out in favor of area specialists.

Given this, we will see that GAI facilitates nearly each one of many above factors. It could possibly generate code, evaluation of knowledge units, and outcomes of queries instantly from textual content prompts. The requirement for AI-ready professionals who can use ChatGPT has already began creeping into job descriptions and we all know that regardless of the productiveness positive factors that include utilizing GAIs, the AI continues to be vulnerable to hallucinations, it might nonetheless get it mistaken, reinforcing the necessity for deep area experience to handle these situations. In abstract, it hasn’t taken 10 years, it’s solely taken two.

Nevertheless, information scientists turning into extinct doesn’t imply people doing information science will develop into out of date, fairly the alternative in truth. After we look again in historical past, during the last 200 years we’ve witnessed a number of technological revolutions, these have included the introduction of steam energy, mass manufacturing, and private computing to call a number of. Every one has enabled us to be extra productive than the final as our roles and relationships with expertise advanced, this idea is properly rooted in financial idea (Solow Progress Mannequin). Within the present atmosphere, companies are creating and capturing extra information than ever, thus information science abilities will at all times be in demand however the information scientists of the long run gained’t be referred to as information scientists, they are going to go by names like product managers, advertising and marketing specialists, or funding analysts. Knowledge scientists are extinct, lengthy reside information science. 


Disclaimer: Views and opinions are the creator’s personal.

Michael Wang is an funding and information science practitioner with over 10 years of business expertise throughout varied roles inside fintech, investments, buying and selling, and educating. He’s the Principal Guide and Founder at WhyPred, an analytics consultancy that specialises in combining monetary markets experience with AI and Machine Studying.

Extra Free Programs on Giant Language Fashions

Mastering NLP Job Interviews – KDnuggets