Do Not Kid Yourselves -- AI WILL be the Default Code Generator

[NewMedia]

Multiple Tech Industry Leaders, National Agencies all over the world, Military, Community collaborations are working towards the Same Goal:
To Replace Human Programmers, including Human “Designers” of App UX experiences.

Of course, they WILL SUCCEED. Screamingly fast progress in Coding, Testing, Learning, Re-Writing to correct bottlenecks, simultaneous run environment testing and re-tooling for efficiency has BEEN HAPPENING.

In fact, it will revolutionize the very code bases of Servers, Internet protocols, browsers, smart devices. Node.js? Replaced. All the frameworks we depend on? Replaced. Server platforms like Apache, Nginx, IIS, Linux, LiteSpeed, etc, etc. will be replaced with a more universal standard that accentuates the performance and capabilities of what enhanced machine learning models produce.

Likewise, all smart devices and computers will be running with AI designed chipsets and operating systems to extract maximum performance from new AI languages.

The tools available for Public Testing are rudimentary compared to what is running in closed development labs. ChatGPT in some ways is now LESS THAN impressive than when it was first released for the public because people have swamped it with requests for “dangerous/destructive/criminal” code. So, AI PEOPLE are consciously curating and confining the parameters of what ChatGPT will return to We the Public.

But, rest assured, INSIDE ChatGPT, with the safeguards removed that confine the Public testing examples, the developers can see already the dangers of the monster they have created.

Other AI research companies mentioned in this video, particularly for Military, Medical applications have opened the Pandora’s Box with working examples of the Capacity for Total Disruption of our world. We just don’t get to see the rough beasts slouching towards Bethlehem until their impact bubbles out of the labs over the next few years.

Will Your Code Write Itself?

[Max_Saravia]

Programming is not an end in itself, it is a means, a tool. If the hammer and the bolt become intelligent and start to create beautiful furniture, it seems perfect to me that the carpenter can really start to be more creative and not be thinking of repeating the same repetitive tasks over and over again in order to express himself. Because let’s face it, those of us who work writing code, what we do the most is repeat over and over again forms, variables, server actions that query, edit and delete data, etc. And when we don’t do that we are trying to make work a code that someone else wrote in our masterpiece. You have to stop being afraid of progress, I can imagine during the industrial revolution people talking practically the same as they are talking now, that machines are going to extinguish the human being or turn them into their personal slave… because every machine surely needs in some point a human slave…
Code exists only to be able to realize our wants and needs in an efficient way… If a machine does it for you, just buy the damn machine and enjoy life that’s what we’re here for

[NewMedia]

My post is not here to express “fear of progress” in coding tools
Not at all.
It is a post about far MORE than mere coding.
Our progress is marked by nuclear bombs even before we created nuclear fission power plants that drove technological progress and much higher standards of living.

Our progress in aviation technologies enabled weapons deliveries all over the world in record time even as all the while commercial aviation advances enabled us to go home for Christmas or to get far, far out of town for business or pleasure.

Our progress in medical technologies developed vaccines that stopped horrific diseases that killed, crippled, decimated global populations.
Same progress that enabled gain-of-function laboratory modified viruses that have killed, agonized, decimated people and societal structures, indeed, national economies on an unseen scale withing a tiny 2 year period.

The BIG PICTURE is the accelerating power of AI to create what we can not even comprehend – for the good, yes. And “progress” that we cannot escape from no matter how we would wish.

I say this on a Web Development forum and choose to Categorize it in the “Coffee Lounge”.

Oh, yes, by the way, the capacity of AI under concerted development by world class geniuses with massive funding to REPLACE THEMSELVES is PROGRESSING at breakneck speed. You see, We Are Now ALL IN!

On the coding tools & development side, the minor things, our “creativity” will take too long to generate and submit for potential client approval. Most Developers will simply know how to query an AI engine more effectively as Middle-Layer service providers. But, that will be a brief phase in history before we just Future “Google” the Application we want.

In fact, I predict the App a company needs becomes an integral process of choosing a Host. An AI Wizard query process asks questions for you to TickBox first and “intuitively” guides you to see templates, so to speak, and walks you through the app while you feed in database requirements, UI choices, etc.

You will have a mighty fine commercial scale application for your company or client within 15 minutes with all kinks de-bugged. Ready for a Click here for installation.

[Max_Saravia]

I think one of the people who predicted all of this, as well as inventing the word “robotics” and being, in my opinion, the best science fiction writer of all time, was Isaac Asimov.
All this reminds me of one of his most famous short stories “The last question”, it is as you say in the end we will only have to ask a computer how to achieve something or order it to create what we imagine or need.
I wonder if the end will be like Multivac hahaha

[scalaris]

Sounds good to me. We can deliver better products faster.

[NewMedia]

I read all of Isaac Asimov!

If I told you that all of my posts today came from a Chat AI engine would you believe me?

Not too long from now it will be a safe Assumption!

[Max_Saravia]

Hhahaha at the end of the day we are all machines carrying a biological body with a nice pc-brain and a pretty elaborate and simple software .

Mad Max future is coming after this super technological era so exercise and keep that body fit, we never know when we will have to run for our lives, be it because of macchines, zombies, aliens or other humans.

[George]

We are also very impressed about ChatGPT and its usefulness about content and code generation.

So we are now researching for ways to add AI functionality it Wappler as well. Ideas are more than welcome!

Some interesting links:

https://chat.openai.com/chat

https://uxplanet.org/openai-chatgpt-dc035deec596

GitHub - f/awesome-chatgpt-prompts: This repo includes ChatGPT promt curation...

[Max_Saravia]

I have been studying and building some models with Pytorch. In my case, the idea is to train artificial intelligence to answer questions asked by clients of one of the companies I work for that is in the vehicle insurance business. The objective is that they can dispense with human personnel on the phone, being able not only to answer specific questions but even to call potential clients to be able to make a sale. Preparing the data sources is really essential, in my attempts I have managed to get artificial intelligence to answer some basic questions, but when the dialogue became more “human” the answers seemed as if I were talking to an infant or an alcoholic person. So I still have a long way to go to achieve what I want but it really has made it much easier to try and build something with these tools.

I agree ChatGpt is really amazing, it has saved me from thinking about several Christmas greetings this year hahaha and my wife has been using it to write some sales pitches.

Regarding Wappler and AI, I can imagine several scenarios that could be useful both to those of us who use the program and to you as developers:

1. Feed an AI with all the questions and answers from the community (a huge task) in addition to the new documentation that you are preparing, so when someone has a question they can ask a virtual helper first instead of spending valuable time searching through topics.

2. In a more advanced use I imagine an AI that can create simple to complex server actions just by naturally typing what one wants. Even helping to deploy an application with all the necessary configurations depending on the server.

3. A robot that delivers beers to Teodor

[NewMedia]

Awesome says it!

What a great Tutorial Engine!
It learns Wappler, various hosting situations, understands flows, etc. and can make up a new Question Customized Tutorial with Images simulating Wappler going through the paces to step-by-step create a Specific Application according to the Questions Asked.

[stinknox]

I’ve seen quite a bit on chatgpt and how it generates code currently. It’s not going to replace human developers any time soon.

That time is coming, but right now most AI can only effectively copy and paste what it has found on stack overflow.

However once it can write code we will be able to build apps faster, integratation with other projects become easier and over all our software becomes more effective at delivering solutions.

I also can’t wait for self driving cars, most people can’t drive anyway. The future looks awesome

[NewMedia]

I’ve been following what the various well-funded major AI research projects have achieved.
FAR more than “copy and paste what it has found on stack overflow” – way beyond such rudimentary data scraping. What we see on the internet as far as public tools is the tip of large and rapidly growing iceberg. The real mass of AI capabilities is below the public water line.

The goal that some of the AI projects have already impressively achieved is to SCRAP any such human code you find on Stack Overflow type sites. To re-imagine machine languages in the first place – which relied on super smart humans building on the foundational genius of other super smart geniuses.

As for self-driving cars, they will both liberate and enslave us.
Self-driving will quickly be subject to AI controlled traffic patterns that determine When, Where, How Long, Allowable stops & detours your “autonomous” vehicle is able to move. What was once as simple as grabbing your keys and heading out the door for a trip to a grocery store or to go camping or vacationing will be strictly curtailed so as to BENEFIT THE MASSES.

This is how commercial aviation has operated already for decades now. Most private pilots can’t simply take off whenever & wherever they want. Personal transportation both on ground and in the air will be ever more tightly controlled.

Sorry, but I look at patterns and you can always predict where they will grow to impinge Control over Personal Freedom.
AI “breakthroughs” are going to be fun in many liberating ways, yes. But will usher in some previously unimagined changes on our lives simply because we will be so Utterly, Helplessly Dependent on their Help & Maintenance of our economies and sheer daily survival.

Look how many millions of people already accept that their smartphones are not only Indispensable to their jobs, leisure, travel, entertainment, banking, ability to pay for goods, etcetera, but that They Are LOST if their phone is not immediately accessible minute by minute. This is a shocking dependence that has arrived while we were celebrating computers and the internet.

[NewMedia]

My very first question in the Desktop ChatGPT app:

Please write a React App with a New User Registration Form that will send a Confirmation of Registration Email notification on successful new registration.

Response:
Sure, here is a sample React app with a new user registration form that will send a confirmation of registration email notification upon successful registration:

import React, { useState } from ‘react’;
import axios from ‘axios’;

function RegisterForm() {
const [name, setName] = useState(’’);
const [email, setEmail] = useState(’’);
const [password, setPassword] = useState(’’);
const [confirmPassword, setConfirmPassword] = useState(’’);
const [error, setError] = useState(’’);
const [success, setSuccess] = useState(’’);

const handleSubmit = (event) => {
event.preventDefault();
if (password !== confirmPassword) {
setError(‘Passwords do not match’);
return;
}
axios.post(’/api/register’, { name, email, password })
.then(() => {
setSuccess(‘Registration successful. An email has been sent to confirm your registration.’);
})
.catch(() => {
setError(‘Error registering user’);
});
};

return (

Name:
<input type=“text” id=“name” value={name} onChange={(event) => setName(event.target.value)} />


Email:
<input type=“email” id=“email” value={email} onChange={(event) => setEmail(event.target.value)} />


Password:
<input type=“password” id=“password” value={password} onChange={(event) => setPassword(event.target.value)} />


Confirm Password:
<input type=“password” id=“confirm-password” value={confirmPassword} onChange={(event) => setConfirmPassword(event.target.value)} />


{error && <p style={{ color: ‘red’ }}>{error}

}
{success && <p style={{ color: ‘green’ }}>{success}}
Submit

);
}

export default RegisterForm;
This form has fields for the user’s name, email, password, and confirm password. It also has state variables for error and success messages, which will be displayed to the user if there is an error or if the registration is successful.

Upon submission of the form, the form’s handleSubmit function is called. This function first checks that the password and confirm password fields match, and displays an error message if they don’t. If the passwords do match, it sends a POST request to the server’s /api/register endpoint with the user’s name, email, and password. If the request is successful, it displays a success message to the user indicating that the registration was successful and an email was sent to confirm the registration. If the request fails, it displays an error message to the user.

*This is just a basic example

Regenerate response

THAT ABOVE is my very first rudimentary question to the AI

ChatGPT Dec 15 Version. Free Research Preview. Our goal is to make AI systems more natural and safe to interact with. Your feedback will help us improve.
New chat
React App with Registration Form

[NewMedia]

I will NOT keep replying to this thread but I just wanted to show my ChatGPT experience when I added one more question to my first question – let me see your server processing script.

My 2nd Question added a new sentence asking for the server side processing – apparently it just picked up where it left off – now writing the React application on the process side picking up the output of its first script – the form itself.

I revised my 1st Question that gave me React code for a Login Form for New User Registration with a new request to follow –

Please write a React App with a New User Registration Form that will send a Confirmation of Registration Email notification on successful new registration. Please then write the complete Server side files with a Postgresql data table named Registration saving the entries and time of registration

React App:

import React, { useState } from ‘react’;

const NewUserRegistrationForm = () => { const [email, setEmail] = useState(’’); const [password, setPassword] = useState(’’); const [confirmPassword, setConfirmPassword] = useState(’’);

const handleSubmit = (e) => { e.preventDefault(); if (password === confirmPassword) { fetch(’/api/register’, { method: ‘POST’, body: JSON.stringify({ email, password, }), }) .then((res) => res.json()) .then((data) => { if (data.success) { alert(‘Successfully registered! A confirmation email has been sent to your email address.’); } else { alert(‘Error registering. Please try again.’); } }); } else { alert(‘Passwords do not match. Please try again.’); } };

return ( Email: <input type=“email” id=“email” value={email} onChange={(e) => setEmail(e.target.value)} />
Password: <input type=“password” id=“password” value={password} onChange={(e) => setPassword(e.target.value)} />
Confirm Password: <input type=“password” id=“confirm-password” value={confirmPassword} onChange={(e) => setConfirmPassword(e.target.value)} />
Submit ); };

export default NewUserRegistrationForm;

Server side files:

const express = require(‘express’); const app = express(); const bodyParser = require(‘body-parser’); const nodemailer = require(‘nodemailer’); const { Pool } = require(‘pg’);

app.use(bodyParser.json());

const pool = new Pool({ user: ‘my_user’, host: ‘localhost’, database: ‘my_database’, password: ‘my_password’, port: 5432, });

app.post(’/api/register’, (req, res) => { const { email, password } = req.body; const query = ‘INSERT INTO registration (email, password, time) VALUES ($1, $2, current_timestamp)’; pool.query(query, [email, password], (error, result) => { if (error) { res.send({ success: false }); } else { res.send({ success: true }); // send confirmation email const transporter = nodemailer.createTransport({ service: ‘gmail’, auth: { user: ‘my_email@gmail.com’,

[StevenM]

Its been a long while since I posted on the forum but to add to this discussion a year or so on, I wonder how long before Wappler needs to seriously think about AI integration?

This will force others to come out of “stealth mode”. The next 12 months are going to be mind blowing!

Cognition emerges from stealth to launch AI software engineer Devin