Why Talking Models are not going to take your jobs [Math Mondays]

Don't worry

Nov 29, 2022

To learn more about the newsletter, check our detailed About Page + FAQs

To help me understand you better, please fill out this anonymous, 2-min survey. If you liked this post, make sure you hit the heart icon in this email.

Recommend this publication to Substack over here

Take the next step by subscribing here

Lately, I've seen a lot of social media influencers talking about how AI will replace every one.

If you've followed AI, you will recognize this trend. However, this time it has gained some serious traction. AI agents are able to do insane things like generate art, write some pretty impressive code, and even decipher jokes. This article goes over some of the more recent innovations that make these tasks possible.

However, from an AI and engineering perspective, this is a long way away from replacing people. And any decent person worth their salt can tell you that. However, these influencers don't really care. They're just here to stoke your fears, make some noise, and profit from the engagement.

For more such top-tier memes, follow me on Instagram and Twitter.

In this article, I will go over the mathematical reason that these models won’t be replacing you, your artist friends, or the many other jobs that these influencers claim.

A lot of AI is based on this formula. Don’t worry, I’ll explain what this means. Image Source

Key Highlights

How data is stored- To understand how these things work (and why they won’t replace people), let’s first understand how models process inputs. In the simplest case (and is most often), input is stored as a vector (which is a fancy math term for a list). Each element of the vector represents a particular feature. Think of as a feature as a characteristic of your sample. If we were creating a data vector of you, things like height, weight, numberOfGrandmasPunched, likesChocolateMilk, etc would all be features.
Comparing Similarity between 2 data points- Remember, each data point is stored as a vector. To compare similarities, all we have to do is compare the distance between the vectors. That is where the formula above becomes helpful. It is called Cosine Similarity. There are other distance metrics you might want to use, based on your needs. As with every other decision, there is a tradeoff. I’ve covered Cosine Similarity in more detail here, for those interested. For our purposes, the important thing to note is that Cosine Similarity can give us the similarity between 2 vectors.
What does this have to do with Big Models- Reasonable question. To oversimplify a behemoth, the models take your user input and try to generate an answer they think would match it well. The match is determined by the similarity (or the complement, which is called disagreement). They do this very well because of the enormous resources sunk into training them. They’re able to build very fine search spaces based on this, which is not the case for standard models.
Why these models will not take over your jobs (utility)- Now for the bit that these influencers conveniently leave out (or haven’t considered). These models’ pattern match existing data. They can’t work with new inputs. For example, if I created a new library, the vaunted GT-CoPilot would struggle with it (it even struggles with existing libraries). If all you did was create art/scripts from older data, you would never create new characters that reflect the tastes of your contemporary audience.
Why these models will not take over your jobs (engineering)- These models have also gotten some engineering problems that make them unusable at scale. For one these are extremely expensive to run. This makes them impractical at scale. Secondly, they are very weak to perturbations and can be broken very easily. Not good for anything that has to be deployed into the real world. I’ll be doing a more detailed into Github CoPilot soon, so make sure you stay tuned for that.

For those of you interested in AI, I created a video in June about why Google’s Lambda model was not alive. At the time of its creation, the internet was going crazy about the possibility of Google having created a model that was alive. It goes over most of these points in more detail. Check it out below-

I created Technology Made Simple using new techniques discovered through tutoring multiple people into top tech firms. The newsletter is designed to help you succeed, saving you from hours wasted on the Leetcode grind. I have a 100% satisfaction policy, so you can try it out at no risk to you. You can read the FAQs and find out more here. Use the button below to get 20% off for upto a whole year.

Get 20% off for 1 year

Before proceeding, if you have enjoyed this post so far, please make sure you like it (the little heart button in the email/post). I also have a special request for you.

Special Request

This newsletter has received a lot of love. If you haven’t already, I would really appreciate it if you could take 5 seconds to let Substack know that they should feature this publication on their pages. This will allow more people to see the newsletter.

There is a simple form in Substack that you can fill up for it. Here it is. Thank you.

https://docs.google.com/forms/d/e/1FAIpQLScs-yyToUvWUXIUuIfxz17dmZfzpNp5g7Gw7JUgzbFEhSxsvw/viewform

To get your Substack URL, follow the following steps-

Open - https://substack.com/
If you haven’t already, log in with your email.
In the top right corner, you will see your icon. Click on it. You will see the drop-down. Click on your name/profile. That will show you the link.

Click on the circled place to get your URL

You will be redirected to your URL. Please put that in to the survey. Appreciate your help.

In the comments below, share what topic you want to focus on. I’d be interested in learning and will cover them. To learn more about the newsletter, check our detailed About Page + FAQs

If you liked this post, make sure you fill out this survey. It’s anonymous and will take 2 minutes of your time. It will help me understand you better, allowing for better content.

https://forms.gle/XfTXSjnC8W2wR9qT9

I see you living the dream.

Go kill all and Stay Woke,

Devansh <3

To make sure you get the most out of Math Mondays, make sure you’re checking in the rest of the days as well. Leverage all the techniques I have discovered through my successful tutoring to easily succeed in your interviews and save your time and energy by joining the premium subscribers down below. Get a discount (for a whole year) using the button below

Get 20% off for 1 year