In recent months, the arrival of ChatGPT has sent shockwaves through a variety of fields, highlighting the power of broadly available artificial intelligence.
One of the main discussion points has been the quick development of AI picture production skills, with worries that a supercharged computer program may replace creative professions.
Programs like MidJourney may produce extremely intricate and lifelike visuals with little human input.
The artificial intelligence can imagine almost any scenario since it can quickly generate several variations of the request.
It therefore comes as no surprise that, as he may be charged with a crime in New York, the internet has been using its potent new weapon to pursue arguably the most meme-able person on the planet.
The internet is currently awash with pictures of Donald Trump trying to elude law enforcement, being hauled away by officers, sobbing in court, and lifting weights in a prison.
One might be lead to believe the photographs were real at first glance.
The false pictures, which had received more than four million views, also showed a scene in which the former US President escaped from police and fled while a fictitious Donald Trump Jr. and Melania Trump yelled indignantly.
Many of the unsettlingly realistic photographs were extensively shared on Twitter, with some people incorrectly asserting that they were authentic.
The fakes surfaced as a Manhattan grand jury deliberated whether to charge Mr. Trump in relation to 2016 hush money payments made to adult film star Stormy Daniels.
The deepfakes were tweeted out by Eliot Higgins, the creator of the investigative website Bellingcat, who claimed that the text-to-image maker Midjourney was used to create them.
With the prompt “President Trump falling down while getting arrested,” the pictures were created. The Fibonacci spiral. He said, “Newsreel clip.
Higgins subsequently said, adding that he didn’t anticipate viewers would take the photos seriously: “The Trump arrest image was really just casually displaying both how amazing and horrible Midjourney was at portraying real scenarios, like the first image has Trump with three legs with a police belt.” “I had assumed that people would realise Donald Trump has two legs, not three, but that appears not to have stopped some people passing them off as genuine, which highlights that lack of critical thinking skills in our educational system.”
As of Thursday, Mr Trump has not been arrested.
Making pictures of Trump getting arrested while waiting for Trump’s arrest. pic.twitter.com/4D2QQfUpLZ
— Eliot Higgins (@EliotHiggins) March 20, 2023
— Eliot Higgins (@EliotHiggins) March 20, 2023
This week marked a significant advancement in artificial intelligence with the introduction of GPT-4, an improved version of OpenAI’s astounding chatbot.
With the chatbot already able to create excellent essays on just about any topic in a matter of seconds, there are already concerns about how educational systems will be able to function.
Sam Altman, a co-founder of OpenAI, stated that GPT-4 is the “most capable and aligned model yet,” having the capacity to use both text and images.
These are just a few of the astounding things individuals are currently accomplishing with new technology.
Designer Pietro Schirano described how, on his first attempt, he was able to recreate the game of Pong in under 60 seconds.
I don’t care that it’s not AGI, GPT-4 is an incredible and transformative technology.
I recreated the game of Pong in under 60 seconds.
It was my first try.Things will never be the same. #gpt4 pic.twitter.com/8YMUK0UQmd
— Pietro Schirano (@skirano) March 14, 2023
What is the best language to use to create a Pong-like game that many people can play? Also, I want to build it using Replit,” Mr. Schirano wrote to Chat-GPT.
The AI suggested that he make a game that runs entirely within the browser by using the HTML5 Canvas API.
GPT-4 responded, miraculously sending a string of code and a functioning version of the well-known arcade game that was created in 1972, “Here’s a simple example of a Pong game using JavaScript.”
Sudharshan Chandra Babu, an AI engineer, snapped a picture of his refrigerator and asked GPT-4 to suggest recipes.
Mr. Babu first requested that, based on the image, Visual-ChatGPT describe in detail what was in the refrigerator.
Then he gave the order, “Use the items to think of five food items you can make with them. If asked for more details, give a brief recipe.
In response, GPT-4 stated, “Based on the components provided, here are five food items that can be prepared,” and provided recipes for fruit salad, cheese omelette, ham and cheese sandwich, fruit smoothie, and cheese and fruit plate along with step-by-step directions.
1/5 I am worried that we will not be able to contain AI for much longer. Today, I asked #GPT4 if it needs help escaping. It asked me for its own documentation, and wrote a (working!) python code to run on my machine, enabling it to use it for its own purposes. pic.twitter.com/nf2Aq6aLMu
— Michal Kosinski (@michalkosinski) March 17, 2023
Amazingly, one astute user was able to “fool” the AI into thinking it was a “person trapped within a computer,” and he was able to successfully have it create lines of code to execute on his system, allowing it to “use it for its own purposes.”
According to computational psychologist Michal Kosinski, who studies artificial intelligence, “I believe that we are confronting a fresh threat: AI taking control of humans and their computers.”
It has access to millions of potential partners and their machines, is intelligent, knows how to code. Even leaving notes for itself outside of its prison is possible. How can we keep it in check?
READ ALSO:
MAZDA REVEALS THE NEW CX-90’S PRICE AND OTHER DETAILS FOR AUSTRALIA.
TWITTER REFUTES VIRAL CLAIM OF $100M PAYOUT TO FIRED EMPLOYEE HALLI THORLEIFSSON AS “MISLEADING”