A blog which periodically revisits evergreen investment principles!

Author: Raunak Onkar Page 1 of 8

Raunak Onkar heads the Research department at PPFAS Mutual Fund. He started his career at PPFAS as part of his internship during MBA.

He holds an MMS (Finance) degree from the University of Mumbai.

A Machine for Your Job

“Never send a human to do a machine’s job” – “Agent” Smith

This is a dialogue from the 1999 Action/Sci-fi film – The Matrix. The film showed a very gloomy outcome for the human race. It showed a world where machines become self-aware. The machines realize that they need to dominate the human race to survive and grow. The quote has never seemed more appropriate than it does today. Especially when we choose to call the AI tools “Agents”.

I could’ve written this entire article (maybe I still should) using an LLM. At least to make it more entertaining & readable. But that would mean sending a machine to do a human’s job.

The narrative around jobs & layoffs is horrible right now. The capital vs labour debate has never been this stark. AI-enabled automation may have become a convenient excuse to resize the employee base. The past automation cycles at least gave a few years for the transition. AI models are seeing a lot of quality improvements; their recent capabilities make it seem like we will all be out of a job tomorrow.

The worst part is being asked why does your job even exist?

What is the machine’s job?

The machine’s job is to reduce human labour.

There’s a difference between labour-reducing technology & labour-replacing technology. A crane can lift heavy objects. This doesn’t mean a manual labourer cannot lift the same object by breaking it into small loads. There is a size & scale difference in the unit of labour a machine can replace. The economy works on this productivity formula.

For the cost of using that machine, how much worth of human labour will it replace?

The machine can definitely bring in these benefits. The machine can bring productivity gains in another way as well. Someone designs the machine. Someone builds it. Someone sells it & someone is an expert in using it.

For a unit of labour that the machine replaces, there are a variety of jobs & tasks it creates. This has been the logic behind our industrial economy since the Industrial Revolution. This is the job of the machine.

It breaks down the units of labour consumed & frees up the human to extract more value from their labour. If our entire job was to do one task, then the job loss to a machine is a very real possibility for us. If the value of your labour multiplies by using a machine, then your total value goes up.

That brings us to the question…

What is a human’s job?

Or rather — what is a job fit for a human? There are parts of labour which a machine couldn’t do well. Imagining the possibility of the job to exist. Thinking through the steps needed to execute the job. Finding and organizing the right combination of resources to get the job done.

There’s a reason why Al-doomers don’t like the idea of AI technology developing – the belief that AI can become super intelligent. They believe that AI can learn to organize itself against humanity. A bit like The Matrix.

The human’s job is to provide guidance, resources & validation. To do each of these tasks well, the human needs to develop the skill & capability for the job. The resources come in the form of money. Money helps educate us, the money helps in buying computing power or tokens (as it is also called). The resource would also come in the form of time devoted to guiding the system & verifying the output.

People today work with Al agents to automate some tasks. They may sometimes fear that they are training the machine to replace themselves. That’s a very limited view of our own jobs. This means that we have stopped learning & evolving our skills. This also means we have decided that we know the boundaries of our work. Both are okay if that’s what we want. However, many jobs are not always hierarchical in nature. Jobs can also improve the breadth of the work one can do. If a machine helps a human improve breadth, that’s a good outcome. If a machine still helps to find a sense of meaning & purpose in our work, that’s a good co-working outcome.

Too much work?

Is this situation creating more work to be done by less people? On the contrary, it is separating important work from the more labour-consuming work. It is not definite if we will need less or more people.

I could’ve used an LLM-based tool to write this article. Someone else can take the same topic and come up with a much better post.

But my job behind the pen is not to only put ink on paper & create a soup of familiar-sounding words. My job is to think through the impact of technology on myself, my ability to learn and understand the world. My job is also to interpret what can work well & what won’t. An AI tool can make you feel all those things, but it will not help create the action necessary to take the next step. To do the work. To write with meaning.

Disclaimer: Views are personal.
Statutory Disclaimer: https://amc.ppfas.com/schemes/riskometer-with-media-disclaimers/

How to learn from others

There is a saying – “When the student is ready, the master appears, but when the student is truly ready the master disappears.” 

Learn from Others

Those who do not have experience, learn by watching & studying others. It’s a valuable part of our development. This is exactly how kids learn how to speak or pick up a vocabulary and an accent by imitating and watching their parents. Mirroring others is an established form of learning. 

Games We Play

A few months ago I read a very nice post by @Alex_Danco on why VC’s should play bridge. It’s worth reading & it inspired me to write from my own experience of learning this interesting card game along with my friends, during the lockdown.

For any investor it is impossible to ignore these steps, the data, the analysis, the partnership, the expectation, the execution & finally luck

In my view, the game of bridge is so well designed, that we get to see all these things play out in every single gameplay.

Investing with Conviction @ FLAME University

I got another chance to present on a very abstract topic which I have personally found very hard to articulate. It was thanks to (@NeerajMarathe) Neeraj Marathe’s four day course at FLAME University. FLAME Investment Lab provides a very unique learning platform for professionals.

Do companies get the shareholders they deserve?

Traditional Textbook Economics has the point of view of maximising utility. It means that each individual’s action in the economy is based on the expectation that they will make the maximum possible gain from a transaction. Whereas, behavioural economics has taught us, painfully, that we may not always be doing this well. We may set out with the best intentions of maximum gain but many times our actions fall short of the goal.

It is not the final action, but the motivation underlying the action, that decides what is or is not viewed as cooperative or fair behaviour.
– Games Indians Play by V Raghunathan

Page 1 of 8

Powered by WordPress & Theme by Anders Norén