90% of online content could be ‘generated by AI by 2025,’ expert says

[Tiffany]

Star Trek and TSNG has always been ahead of the curve ball in terms of seeing the future, or at least the writers were very imaginative and really brought to us advanced future probabilities to consider.

 

[Crims]

Has anyone tried programming with ChatGPT? It seemed ok for coding Unity and Unreal.

 

[Arantor]

I have for PHP. Results were predictably abysmal - to the point where it was giving me code that flat out wouldn't work and had multiple security holes. But the stochastic parrot wouldn't know any different because I guarantee wherever it scraped the code from wouldn't have known any better either.

 

[Crims]

That's kind of good. I'm hoping that they'll eventually be capable of basics... but the idea that a regular of the language would do better is a relief.

 

[Arantor]

That's really the thing: people talk about how amazing the thing is but it's really still a stochastic parrot, repeating what it has in its bowels, funnelled through a maze of statistical modellings until it thinks it's arrived at something that matches.

It has no comprehension, no matter what the advocates say, and it has absolutely no awareness of whether what it is saying is *correct*. If you haven't seen it, https://www.digitaltrends.com/computing/chatgpt-bing-hands-on/ is a fantastic quick study into the descent into madness that these things can have precisely because it has no awareness of anything.

 

[Arizona]

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies

"We build a Generatively Pretrained Transformer (GPT), following the paper "Attention is All You Need" and OpenAI's GPT-2 / GPT-3. We talk about connections to ChatGPT, which has taken the world by storm. We watch GitHub Copilot, itself a GPT, help us write a GPT..."

 

[Arizona]

I asked ChatGPT (through Microsoft's Bing chat) for Python code for the KMP algorithm.

The answer is better commented than most human answers I have seen:

Here is an example of the KMP algorithm implemented in Python:

def KMPSearch(pat, txt):
M = len(pat)
N = len(txt)

# create lps[] that will hold the longest prefix suffix
# values for pattern
lps = [0]*M
j = 0 # index for pat[]

# Preprocess the pattern (calculate lps[] array)
computeLPSArray(pat, M, lps)

i = 0 # index for txt[]
while i < N:
if pat[j] == txt:
i += 1
j += 1

if j == M:
print("Found pattern at index " + str(i-j))
j = lps[j-1]

# mismatch after j matches
elif i < N and pat[j] != txt:
# Do not match lps[0..lps[j-1]] characters,
# they will match anyway
if j != 0:
j = lps[j-1]
else:
i += 1

def computeLPSArray(pat, M, lps):
len = 0 # length of the previous longest prefix suffix

lps[0] = 0 # lps[0] is always 0
i = 1

# the loop calculates lps for i = 1 to M-1
while i < M:
if pat == pat[len]:
len += 1
lps = len
i += 1
else:
if len != 0:
len = lps[len-1]

# Also, note that we do not increment i here
else:
lps = 0
i += 1

txt = "ABABDABACDABABCABAB"
pat = "ABABCABAB"
KMPSearch(pat, txt)

This code creates two functions: KMPSearch and computeLPSArray. The KMPSearch function takes in a pattern and a text string as arguments and searches for occurrences of the pattern within the text string. The computeLPSArray function takes in a pattern, its length, and an array to hold the longest prefix suffix values for the pattern. It preprocesses the pattern to calculate the values in the array.

 

[CyberGeek]

This AI is dangerous. If misinformation is a problem now, it will be nothing compared to computerised comments posted everywhere, pushing an agenda, propaganda.

You can be sure that governments, political parties and dictators will misuse this technology for their own benefit. We might see regulations come in eventually over its use, but it won't be enough.

Just imagine having a heated argument with a troll in an article comment section only for that to be AI, but hiding that fact. And that's only the start.

The thought of the last paragraph made me laugh

 

[CyberGeek]

I wonder how the AI feels about being profiteered off?

I'm joking now, but if such an AI ever becomes sentient, then it may well have something to say about this.

How would Meghan have felt being sold off like a slave?

Anyone who’s seen the movie M3ghan will get the reference !

 

[CyberGeek]

I keep thinking of Skynet every time I read the above posts about AI

I get paywalls and login to read this article but also find them annoying. I know what they are doing and why, but what annoys me is some websites take information from other websites, put a twist of their own on the article and paywall it. I've found that I can take the same title of the website that's pay-walled their article, and find the same information on another website that's free.

I can relate to this FFS

 

[Arizona]

What I find most impressive about ChatGPT for coding is that it is familiar with a number of third-party libaries.

Two example code responses from ChatCPT, one with the popular Wordpress API and the second with a less-known library for visualization called Nivo.rocks:

Here’s an example of how you can use the WordPress REST API to retrieve all posts and save them to an SQL database:

import requests
import json
import pymysql

# Connect to the database
connection = pymysql.connect(host='localhost',
user='your_username',
password='your_password',
db='your_database_name')

# Create a cursor object
cursor = connection.cursor()

# Set the URL for the WordPress REST API endpoint
url = 'https://yourwordpresssite.com/wp-json/wp/v2/posts'

# Make a GET request to the WordPress REST API
response = requests.get(url)

# Load the response data as a JSON object
data = json.loads(response.text)

# Loop through each post in the data
for post in data:
# Get the post ID, title, and content
post_id = post['id']
post_title = post['title']['rendered']
post_content = post['content']['rendered']

# Create an INSERT statement to insert the post into the database
sql = "INSERT INTO your_table_name (post_id, post_title, post_content) VALUES (%s, %s, %s)"

# Execute the INSERT statement
cursor.execute(sql, (post_id, post_title, post_content))

# Commit the changes to the database
connection.commit()

# Close the database connection
connection.close()

This code uses Python’s requests library to make a GET request to the WordPress REST API and retrieve all posts. The posts are then saved to an SQL database using Python’s pymysql library. You’ll need to replace your_username, your_password, your_database_name, yourwordpresssite.com, and your_table_name with your own values.

Sure! Here’s an example of how you can create a bar chart using React and Nivo.rocks:

import React from 'react';
import { ResponsiveBar } from '@nivo/bar';

const data = [
{
"country": "AD",
"hot dog": 134,
"hot dogColor": "hsl(183, 70%, 50%)",
"burger": 12,
"burgerColor": "hsl(183, 70%, 50%)",
"sandwich": 190,
"sandwichColor": "hsl(183, 70%, 50%)",
"kebab": 184,
"kebabColor": "hsl(183, 70%, 50%)",
"fries": 16,
"friesColor": "hsl(183, 70%, 50%)",
"donut": 68,
"donutColor": "hsl(183, 70%, 50%)"
},
];

const MyResponsiveBar = () => (
<ResponsiveBar
data={data}
keys={[ 'hot dog', 'burger', 'sandwich', 'kebab', 'fries', 'donut' ]}
indexBy="country"
margin={{ top: 50, right: 130, bottom: 50, left: 60 }}
padding={0.3}
valueScale={{ type: 'linear' }}
indexScale={{ type: 'band', round: true }}
colors={{ scheme: 'nivo' }}
borderColor={{ from: 'color', modifiers: [ [ 'darker', 1.6 ] ] }}
axisTop={null}
axisRight={null}
axisBottom={{
tickSize: 5,
tickPadding: 5,
tickRotation: 0,
legend: 'country',
legendPosition: 'middle',
legendOffset: 32
}}
axisLeft={{
tickSize: 5,
tickPadding: 5,
tickRotation: 0,
legend: 'food',
legendPosition: 'middle',
legendOffset: -40
}}
labelSkipWidth={12}
labelSkipHeight={12}
labelTextColor={{ from: 'color', modifiers: [ [ 'darker', 1.6 ] ] }}
legends={[
{
dataFrom: 'keys',
anchor: 'bottom-right',
direction: 'column',
justify: false,
translateX:120,
translateY:-10,
itemsSpacing:-10,
itemWidth:-10,
itemHeight:-10,
itemDirection:'left-to-right',
itemOpacity:-10,
symbolSize:-10,
effects:[
{
on:'hover',
style:{
itemOpacity:-10
}
}
]
}
]}
animate={true}
motionStiffness={90}
motionDamping={15}
/>
)

This code creates a bar chart using the ResponsiveBar component from the @nivo/bar package. The data for the chart is defined in the data variable and passed to the data prop of the ResponsiveBar component. You can customize the appearance and behavior of the chart by adjusting the props passed to the ResponsiveBar component.

I hope this helps! Let me know if you have any questions or need further assistance.

 

[CyberGeek]

People deserve to be paid for their work, indeed....my point was when large websites copy original work and paywall it as if the news is exclusive to them (eg. the website copies the work without acknowledgement to the original author), that's where I get annoyed...I think that's dishonest. I wasn't too clear.

Cute on the ChatGBT for Nerdzone!

Name a human who is genuinely honest on planet Earth.

 

[CyberGeek]

What I find most impressive about ChatGPT for coding is that it is familiar with a number of third-party libaries.

Two example code responses from ChatCPT, one with the popular Wordpress API and the second with a less-known library for visualization called Nivo.rocks:

Oh come on FFS that’s just regurgitating the documentation !

Can it distinguish between fact and opinion? No, clearly not, like the example given about visiting this forum regularly.

What if some crooks endorsed some illegal activity on the internet . ChatGPT will gather that info and respond to the question “How often should we commit crimes?” with “We should commit crimes regularly.”

 

[CyberGeek]

I read somewhere that chatGPT can explain what code does.I recommended this solution to my boss and he was concerned about the guesswork based on other solutions and the inability to determine novel solutions

 

[Arizona]

Oh come on FFS that’s just regurgitating the documentation !

But within seconds and with great comments (sometimes).

 

[CyberGeek]

But within seconds and with great comments.

Yeah I know I posted but within seconds and with great comments

 

[CyberGeek]

But within seconds and with great comments.

The speed is impressive, sure, but with web scraping, web crawlers, search engines and all that that’s to be expected

 

[Arizona]

To me, it's a similar situation with students writing a term paper. ChatGPT will help get the ball rolling and with providing information/education. For coding, this may help free one's mind from trivial details to more design concerns.

It may also help novice and mid-level developers to understand new frameworks, reducing "cognitive load."

It is familiar with the Xenforo forum API as well:

Here’s an example of how you can get the 10 most recent threads from XenForo using the XenForo API:

<?php
// Set your API key and forum URL
$apiKey = 'YOUR_API_KEY';
$forumUrl = 'https://yourforum.com';

// Create a function to make requests to the XenForo API
function xf_request($method, $endpoint, $data = array()) {
global $apiKey, $forumUrl;

// Set the request URL
$url = $forumUrl . '/api/' . $endpoint;

// Set the request headers
$headers = array(
'XF-Api-Key: ' . $apiKey,
'Content-Type: application/json'
);

// Initialize cURL
$ch = curl_init();

// Set the cURL options
curl_setopt($ch, CURLOPT_URL, $url);
curl_setopt($ch, CURLOPT_HTTPHEADER, $headers);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);

// Set the request method and data
if ($method == 'POST') {
curl_setopt($ch, CURLOPT_POST, true);
curl_setopt($ch, CURLOPT_POSTFIELDS, json_encode($data));
} else if ($method == 'GET' && !empty($data)) {
curl_setopt($ch, CURLOPT_URL, $url . '?' . http_build_query($data));
}

// Execute the cURL request
$response = curl_exec($ch);

// Close cURL
curl_close($ch);

// Decode the response and return it
return json_decode($response);
}

// Get the 10 most recent threads using the XenForo API
$threads = xf_request('GET', 'threads', array('limit' => 10));

// Print the threads
print_r($threads);

This code is similar to the previous example but adds an additional parameter to the xf_request function call to specify that only 10 threads should be returned. This is done by passing an array with a limit key set to 10 as the third argument to the xf_request function.

The response from the API will contain only the 10 most recent threads. These are decoded from JSON and printed.

 

[CyberGeek]

To me, it's a similar situation with students writing a term paper. ChatGPT will help get the ball rolling and with providing information/education. For coding, this may help free one's mind from trivial details to more design concerns.

It may also help novice and mid-level developers to understand new frameworks, reducing "cognitive load."

It is familiar with the Xenforo forum API as well:

Yeah the concern my boss had was “what if it’s inaccurate? ChatGPT is relatively new so perhaps with time that view will change. But there ‘s always a chance that AI could be wrong and that could F things up.”

 

[Arizona]

For detecting AI content:

New AI classifier for indicating AI-written text

We’re launching a classifier trained to distinguish between AI-written and human-written text.

openai.com

OpenAI Platform

Explore developer resources, tutorials, API docs, and dynamic examples to get the most out of OpenAI's platform.

platform.openai.com

The AI Text Classifier is a fine-tuned GPT model that predicts how likely it is that a piece of text was generated by AI from a variety of sources, such as ChatGPT.

This classifier is available as a free tool to spark discussions on AI literacy. For more information on ChatGPT's capabilities, limitations, and considerations in educational settings, please visit our documentation.

 

[Arizona]

OMG! The arrogance of the OpenAI presenter just now, talking about the new version, ChatGPT4:

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies

"It's not perfect, but neither are you."

~ Greg Brockman, President and Co-Founder of OpenAI

 

[Arizona]

Introducing Genai - Bringing ChatGPT To Interactive Python

genai is an IPython extension that brings ChatGPT to help you diagnose errors and exceptions quickly. The motto is Errors Make Perfect. Keep iterating to learn and grow! Jupyter notebook.

noteable.io

"An open-source library that brings the power of ChatGPT into error handling within Jupyter notebook."

 

[Retro]

There's a new version out, GPT-4.

OpenAI announces ChatGPT successor GPT-4

The fourth version of the AI chatbot can process both images and text.

www.bbc.co.uk

 

[Tiffany]

Thanks for all of the info....lots to read and videos too.

 

[Arizona]

Will AI Advancement Mean the End of Programming?

AI seems to be making huge strides every single day. As a programmer, should you be worried that the machines are coming for your job?

www.makeuseof.com

"A single neuron in the brain is an incredibly complex machine that even today we don’t understand. A single ‘neuron’ in a neural network is an incredibly simple mathematical function that captures a minuscule fraction of the complexity of a biological neuron."

~ Andrew Ng