But first, let’s start with some perspective. If we glance back to the introduction of business intelligence tools within the early 2000s, the nice value of those tools lies of their ability to offer non-technical, line-of-businesspeople the power to leverage their domain knowledgeby enabling them to pick, analyze and present data, without writing a stitch of code. Sound familiar?

Providing user-friendly means to research data is nothing recent. It can all the time have incredible value. Indeed, it’s a multi-billion dollar industry that continues to grow. Nonetheless, these tools don’t have any use without . This is applicable to any data evaluation, whatever the tool(s) getting used. Even when it’s generative AI. Without domain knowledge, we have no idea what inquiries to ask of our data. And even when the questions were provided to us, how will we interpret our findings?

And for my part, the best value of knowledge evaluation work lies in its ability to reply ad hoc questions. Unexpected, mission-critical questions. Complex, multi-layered, nonlinear forms of questions. Answering these questions requires domain knowledge.

For instance, why did sales on our best-selling product just drop off a cliff? Our primary supplier just went out of business, what will we do? Why did our customer churn rate double last month? These aren’t straightforward forms of questions that may follow a longtime decision tree.

What these few examples have in common is that they require immediate answers to situational questions which have never been asked before. And that is absolutely the important thing. In the event you understand the construct of generative AI, its inability to reply questions of this nature is actually its Achilles heel in ever having the ability to replace data analysts fully.

To briefly summarize, generative AI utilizes existing data sets to ‘train’ an LLM to a probability-driven answer based on whatever training data it has been fed. And while you possibly can constantly fine-tune your model with ever more precise data sets, how would you train your model on multi-layered, situational questions which have never been asked before?

It will be analogous to you starting a recent job as a knowledge analyst in an industry that you just aren’t yet acquainted with. And on day one, you’re asked to urgently answer one in all the questions above. Where would you even start? What data would you pull? How would you even know what all the potential variables you would want to think about? And, even for those who could in some way derive a solution, how would whether it is correct?

It’s for these reasons that I don’t foresee the role of knowledge analyst ever being fully replaced by generative AI. Nonetheless… generative AI, in its current state, already has many uses in the information evaluation field and people uses will only proceed to expand with ever-increasing functionality.

Current Potential Uses for Generative AI in Data Evaluation

As of today, the very best and best use of generative AI in the information evaluation field is its ability to each write code and in turn, explain the code it writes (which it does quite well). I’ve personally used it to assist me write and understand Python code.

For those of you who want to enter the information evaluation field, I couldn’t encourage you sufficient to make the most of generative AI to show you how to learn to code. It will have greatly accelerated my learning curve once I was first cutting my teeth on this field.

In one other, truly exciting development for data analysts, generative AI has fueled the event of dedicated coding tools. GitHub has released its product, which might suggest coding solutions/improvements in real-time as you’re writing it!

Earlier in this text, I referred to the potential hurdles firms would face in constructing their very own LLMs. There’s possibly one recent alternative to that: Databricks has recently released an LLM called ’ ’. In theory, this might solve the problems of cost (being open source) and having to push your data outside of your organization’s firewall. It’s a smaller-scale LLM, more fitted to focused datasets.

I mention Dolly, primarily for instance of how quickly developments in the sector of generative AI are moving and as a heads-up to how they could affect the information evaluation field going forward.

As we now have already seen, the evolution of AI will only proceed to progress at light speed.

Conclusion

There isn’t any doubt in my mind that generative AI will reshape the workflows in data evaluation. Generally speaking, repetitive forms of tasks and even analyses will in time be performed by generative AI. I could also see coding becoming more of a commodity, versus being a highly developed skill.

Based on the above, I feel that the prototypical data analyst in the long run will possess business line-level domain knowledge combined with a capability to include generative AI tools to assist them be more efficient and productive with their time.

Lastly, on a private note, I’d encourage anyone reading this to embrace generative AI. Find out about it and use it each in your personal and business lives. With recent APIs and plugins always being created, its reach and capabilities will only grow.

For higher or worse.