The writer of that article, assuming they did the graphics, are...

The writer of that article, assuming they did the graphics, are not very good at maths. 35Gw to 122 Gw = triple, close, but 3.5 times is closer!! plus when you add....
" there are several other large-scale facilities coming online in the next five years that will potentially bring that number substantially higher."

We are going a lot higher than triple in 5 years, closer to quadruple or higher. This is a growth rate of over 40%/a for the next 5 years, if the other sources don't happen, and closer to 50%/a if they do.

Despite the planned factories, I've seen analysts predict 15-20% growth for lithium, that is clearly wrong.

This can only mean that lower grades of lithium NEED to be used to get the quantities to match demand. For lower grade resources, they will only be developed if it is profitable, therefore the price must rise a lot in a very short period of time.
My opinion is that we could clearly work with a price of $1000/t of 6% spodumene, with the cost of high grade operations being around $90/t (from AJM scoping study in 2012). That means the resource we have is worth way MORE than most think!!

IE at 52mt of 1.25% = 13mt of 6% =$13b with $11.7b after costs = $over 10/sh... gross PROFIT

Now if the price doubles again........