We are talking about AI and think it is all
awesome and stuff.
It is.
BUT!
It does not mean we will work less, it means we will work the same and get
more done. It also means that people who locked out of certain industries will move i to them.
In the mid 1800’s the six shot revolver was developed. It changed who and how close quarters combat was conducted. Prior to the revolver, swords and upper body strength ruled. After, skill and speed ruled. It changed so much that the revolver was called the great equalizer.
I encourage everyone to watch this video
This is 26 minutes and he pours out information like a firehose. It is all
practical so may be somewhat boring to some. I recommend you
watch it anyway. Not because you need to learn to be an AI wrangler, but so that you see just how complex this is going to become.
I suspect that AI is more than going to transform our companies, it is
going to create not just new companies, but new industries.
The idea of script crafting and using AI to script craft for other AI’s is going to be a business. I have no idea who is going to
make money off of this, nor how, but a at least passing knowledge of how this is going to be creatively used will help us parse out the silly from the substantive.
See my silly example.
Please write the code for “Hello World” in the programming languages: Fortran, C , Python
Sure, here are the “Hello World” programs in Fortran, C, and Python:
Fortran:
fortranCopy code
program hello print *, "Hello, World!" end program hello
C:
cCopy code
#include <stdio.h> int main() { printf("Hello, World!\n"); return 0; }
Python:
pythonCopy code
print("Hello, World!")
Note: For the Python example, you can run it in any Python environment or even in a Jupyter notebook.
The next prompt is this:
Can you show the compiled C code in hexadeximal and binary?
Yes, I can show you the compiled C code in hexadecimal and binary. Here’s an example code in C that prints “Hello, World!” and its corresponding compiled code:
C code:
cCopy code
#include <stdio.h> int main() { printf("Hello, World!\n"); return 0; }
Compiled code (x86_64 architecture, Linux):
Copy code
55 48 89 e5 48 83 ec 10 bf 14 00 60 00 b8 01 00 00 00 48 89 e6 ba 0d 00 00 00 bf 01 00 00 00 b8 04 00 00 00 0f 05 b8 00 00 00 00 c9 c3
Here’s the same compiled code in hexadecimal:
Copy code
55 48 89 e5 48 83 ec 10 bf 14 00 60 00 b8 01 00 00 00 48 89 e6 ba 0d 00 00 00 bf 01 00 00 00 b8 04 00 00 00 0f 05 b8 00 00 00 00 c9 c3
And in binary:
Copy code
01010101 01001000 10001001 11100101 01001000 10000011 11101100 00010000 10111111 00010100 00000000 01100000 00001000 10111000 10001001 11100110 10111010 00001101 00000000 00000000 10111111 00000001 00000000 00000000 00000000 01001000 10001001 11100110 10111000 10111111 00000001 00000000 00000000 00000000 10111000 00000100 00000000 00000000 00001111 00000101 10111000 00000000 00000000 00000000 00000000 11001001 11000011
Note that the compiled code can vary depending on the system and architecture it was compiled on
Now I am not that smart, or
maybe I am, I would rather be sailing, so how to use this power to say have the AI
write the code to pull all the quarterly reports from the Fred database and sort it for false information.
See the next AI Prompt.
There is a math mathematical rule that says if a report has a certain number of nines which are less common than ones, that the data is probably false. What is that rule?
The rule you’re referring to is called Benford’s Law, named after physicist and mathematician Frank Benford who observed this phenomenon in 1938. Benford’s Law states that in many naturally occurring collections of numbers, the leading digit is likely to be small. In fact, the frequency of occurrence of the leading digit d is given by:
log10(1 + 1/d)
So, for example, the probability of the leading digit being 1 is about 30%, and the probability of it being 9 is less than 5%.
If a set of numbers doesn’t follow Benford’s Law, it could indicate that the data has been manipulated or is not naturally occurring. However, it’s important to note that Benford’s Law is not foolproof and should be used with caution. There may be legitimate reasons why a dataset doesn’t follow the law, and it’s always important to investigate further before drawing any conclusions.
So can you write a python code to pull that last 12 quarterly reports and the last the last three annual reports of all of the companies in the Fred database so that they can be analyzed with Benfords law?
Sure! Here’s a Python code that uses the FRED API to fetch the last 12 quarterly reports and the last 3 annual reports for each company in the FRED database:
python
Copy code
import requests
import json
Replace YOUR_API_KEY with your actual API key from FRED
api_key = “YOUR_API_KEY”
Endpoint for getting company data
url = “https://api.stlouisfed.org/fred/category/children”
Parameters for getting quarterly reports
params_qtr = {
“category_id”: “32893”,
“api_key”: api_key,
“file_type”: “json”
}
Parameters for getting annual reports
params_ann = {
“category_id”: “32908”,
“api_key”: api_key,
“file_type”: “json”
}
Get the list of companies from the FRED database
response = requests.get(url, params=params_qtr)
data = json.loads(response.text)
companies = [child[‘id’] for child in data[‘categories’][0][‘childcategories’]]
Fetch the last 12 quarterly reports and last 3 annual reports for each company
for company in companies:
# Quarterly reports
response_qtr = requests.get(f"https://api.stlouisfed.org/fred/category/series?category_id={company}", params=params_qtr)
data_qtr = json.loads(response_qtr.text)
quarterly_reports = [series[‘id’] for series in data_qtr[‘seriess’][-12:]]
# Annual reports
response_ann = requests.get(f"https://api.stlouisfed.org/fred/category/series?category_id={company}", params=params_ann)
data_ann = json.loads(response_ann.text)
annual_reports = [series['id'] for series in data_ann['seriess'][-3:]]
# Process the data using Benford's Law
# ...
# Your code for analyzing the data with Benford's Law goes here
# ...
Note that you’ll need to replace “YOUR_API_KEY” with your actual API key from FRED. Also, the code for analyzing the data using Benford’s Law is not included, as it depends on the specific method you’re using to analyze the data.
Ok, at this point I would need to get off the couch and move from using the Iphone mini to a real computer with Python loaded. This seems like work so I will stop here.
Also, I suspect that is you attempted to run this code you would need a super fast connection and many terabytes of storage. Then you would have to have the AI write code to clean the data, then write
code analyze the data. .
Cheers
Qazulight (Watch the video!)