Why Karl Marx would have loved Artificial Intelligence. Or how to empower BI analysts.

Karl Marx.jpg

You know that feeling you have when you’ve had a long hard day of work and you’re eventually in bed. Nice crisp, clean sheets. Reading a good book. Clearing your head with something totally different than what you’ve spent the day on.

Well, I was there yesterday evening, exactly at the point where I was to get into the arms of Morpheus. Eyelids starting to get heavy while reading “Great thinkers”. A great book that presents the work of political thinkers, philosophers and other great thinkers, a couple of pages for each of them. Highly recommend if you want to understand, in a few minutes, what great thinkers from Aristotle all the way to Adam Smith via Sartre and Schopenhauer said.

I was actually reading Karl Marx and one of his critics of capitalism. And in a very weird way, I could not help to think about the parallel between what Karl Marx was saying and what I was seeing on a daily basis while taking to market PredicSis.ai, a self-serve analytic platform using artificial intelligence technology.

 

Hmm? Excuse me!!? Karl Marx? Artificial Intelligence?

Yeap, you read this one well. I could not help to see an obvious link between Karl Marx main analysis of the issues of capitalism brought to workers of his time (and ultimately ours) and the opportunities that AI brings. But let me explain and get some specifics so you understand where I come from.  

 

Karl Marx for dummies

For those of you who haven’t read Das Kapital (or those who haven’t read the two pages in the book I mentioned above), one of Karl Marx major criticism of the capitalist system is that it prevents people from expressing themselves fully. In Karl Marx views, a poet would want to have a go working in a factory for a short while or an accountant would relish the opportunity to work as a landscape gardener.  On the contrary, the capitalist system drives organisation of labour towards specialism, towards a single task for efficiency’s sake. This made modern workers alienated.

 

So, there is a conflict between what work should be, a source of fulfilment where people can express themselves and what it really is, i.e. a very specialised field. Ergo, the specialisation is essentially an economic imperative but is a human betrayal (again, read the Great thinkers book even if you haven’t!).

Data analysts & AI making Karl Marx a happy man?

Driving a company towards a customer centric and data driven approach can be seen as difficult and lengthy. It is a transverse process, based on a mix of individual data in order to segment, predict and act ideally automatically.  

 

Data Analysts are the ones who need to enhance organisation’s performance through the use of data. They need to make sure the reasons of fraud are fully understood and fraud kept at its lowest. They are in charge of helping sales and marketing to reduce churn. Their knowledge of “the data” and what it means in the context of their business is critical and used to drive efficient upsell and cross-sell campaigns. They are looking into datasets to analyse clickstream and drive adoption of a service. The list goes on and on. These professionals are key to understand how an organisation does make sense and how it can improve its existing process.

 

In a classic organisation, dare I say a “pre-AI organisation”, a data analyst core function is focused in finding meaning into various datasets. As data sources are expanding, should she need to drive predictive projects, she relies on a data scientist team to produce models to make sense of it. She does a lot of data preparation which is a rather tedious task (Karl Marx anyone?). Then passes these to the data scientist(s) who would work out some modeling. Then she would get a model back from the data science team, make some comments on what need to be changed and a back and forth process would ensue. The strategy team would get involved somehow to get their views on what the insights mean. Then, she would hand over to DevOps to implement this model in a production environment. Taylorisation would have done its toll, breaking down the job of the data analyst in small parts.

 

In a post-AI organisation, the taylorisation of the analyst job can be taken apart. She could look through the data with her business stakeholders without facing “the big data wall”, the “R or Python” hurdle, the “that won’t pass production” effect and not even the “my data mart is not designed to facilitate customer centric queries”. She can inject her data, small or large in an predictive platform (thanks to solutions like PredicSis.ai), just dragging some imperfect csv files and get a view of what the data means, in minutes. And even work closely with the devOps team to implement this model in real life.

 

In short, AI would be able to let a data analyst express a much wider perspective of her personality at work, to interact with various stakeholders around what the data means and how to use it. She would be able to gather and share strategic perspective or get her teeth into the technicalities of datasets and the deployment of models in real life for A/B testing.

 

I know, this is taking us far far away from the nice feelings that one can have by reading his book while laying in nice crisp sheets. And I usually try to steer away as much as possible from the AI word. But, as the debate around AI is starting to take off across all spectrums of society, I could not help but to see how the analysis made by Karl Marx, who were very true and appealing to many, would not necessarily stand the test of AI in some parts of our economy.

 


Intrigued? If you want to get a sense of PredicSis.ai, we made it easy for you through our demo page