Tesla stock price in 2026

What if AI got really good at predicting stocks?

What you see above is the forecasted stock price for Tesla for the next 3 years predicted with AI.

Though I have a feeling the stock is going to swing more wildly like we see with the historical data.

Let’s get into it

Here’s what I learned:

• Code Interpreter (OpenAI) can create graphs (& honestly it’s just the tip of the iceberg…)

• It can analyze financial data (and data of any kind)

• It can try to predict the future if you prompt it

Warning: this isn’t perfect - it’s a prediction made by some early stage AI model so treat it as such. This is not investment advice.

Here’s what I did, step-by-step

I went to Yahoo Finance:

• Got this historical data for Tesla from the last 5 years

• Downloaded that CSV

• Uploaded that CSV to Code Interpreter

After that I prompted the following:

• “I'd like for you to give me some graphs and infographics of this data for the 5 years, 1 year, last 6 months, and more. Be thorough.”

Amazingly, it created and designed the graphs below:

Tesla Adjusted Close Price - Last 5 Years

Trading Volume - Last 5 Years

Adjusted Close Price - Last 1 Year

Pretty good! And these aren’t even all of the graphs it generated…

(As an aside, let me know if there are data you’d like me to analyze because there’s soooooo much you can do - expect more from me on this.)

So after that, I followed up with the following prompt:

• “Now given this data, I'd like for you to predict what the next 5 years will look like and provide visualizations as well. I know that it's very uncertain and that you won't be correct but just please predict and use the best of your ability. This is just an exercise. Thanks!”

What I did here was make sure it was comfortable giving me a prediction.

A lot of the time, the model doesn’t want to give you a prediction because it doesn’t feel comfortable doing so…

So you want to weed that out early on.

At first it wanted to use the “AutoRegressive Integrated Moving Average” (ARIMA) model.

It performed a series of tasks and gave me the following:

This was a flop…

So I went back to the prompt and literally said:

• “It's just a flat line, I need real predictions here! Do it again!”

I know, I was was kinda mean (but I wanted to see how it’d react).

It had explain that that method didn’t work for some reason, so it tried another.

Final prompt

I wanted to narrow it down just a little bit more - make it work just a bit less so I made the parameters 1-3 years instead of 5.

I asked it to:

• “Focus on 1-3 years using the new model you're suggesting!”

It handled this much better…

Output

• “Let's use the Exponential Smoothing (ETS) method, which is a commonly used statistical approach to time series forecasting. We'll forecast the adjusted close price for the next 3 years using the Holt-Winters Exponential Smoothing method.”

Forecasted Adjusted Close Price - Next 3 Years

Giving us this beautiful graph. It even created a new red line from today to August 2026! So cool.

Well now that I know Tesla will be up, I’m going all in! 💰

Juuuust kidding.

Try this for yourself

• Follow this guide

• Try downloading a dataset from Kaggle or elsewhere

• Upload it to Code Interpreter and make some cool stuff

Let me know what you make!

Know someone who likes data? Send them this newsletter:

daily-build.beehiiv.com 

Need some help implementing AI tools at work?

Hire me