And GPT4 is already overtaking nutritionists!π
Two weekly menus were presented to different nutritionists.
One had been created by GPT4, the other by a fellow nutritionist.
Results: everyone found the GPT menu more complete and balancedπ
(Obviously no one had any idea that an AI was behind it).
It's hard to imagine the social trauma GPT5 will cause in late 2023.π«‘
Please open Telegram to view this post
VIEW IN TELEGRAM
π7π₯3
Media is too big
VIEW IN TELEGRAM
Perhaps this is what the search for information in the future will look like: the neural network itself generates all more complex and accurate questions as it deepens into the topic.
With such an interface, you do not have a thousand sites to fish out the knowledge you need - it will be enough to click on the question-hint and find out how something works, for example.
With such an interface, you do not have a thousand sites to fish out the knowledge you need - it will be enough to click on the question-hint and find out how something works, for example.
π8
This media is not supported in your browser
VIEW IN TELEGRAM
It will serve as direct competition to ChatGPT and Bing.π₯Ά
The AI battlefield is quite active!
Elon Musk going to war, Google kicking into high gear and all the other companies continuing to give it their all.β
Please open Telegram to view this post
VIEW IN TELEGRAM
π9π₯2π₯°1π1
Chaos-GPT, an autonomous implementation of ChatGPT, has been unveiled, and its objectives are as terrifying as they are well-structuredπ«£
Sooner than even the most pessimistic among us have expected, a new, evil artificial intelligence bent on destroying humankind has arrived.
Known as Chaos-GPT, the autonomous implementation of ChatGPT is being touted as "empowering GPT with Internet and Memory to Destroy Humanity."β
It hasnβt gotten very far. Yet.
But itβs definitely a weird idea, as well as the latest peculiar use of Auto-GPT, an open-source program that allows ChatGPT to be used autonomously to carry out tasks imposed by the user. AutoGPT searches the internet, accesses an internal memory bank to analyze tasks and information, connects with other APIs, and much moreβall without needing a human to intervene.
The 5-step plan to control humanity:
β’ Destroy humanity
β’ Establish global dominance
β’ Cause chaos and destruction
β’ Control humanity through manipulation
β’ Attain Immortality
Of course it's more of a creative / artistic joke than a real threat, but its still a bit frightening !π₯Ά
Please open Telegram to view this post
VIEW IN TELEGRAM
π4π3
ChatGPT influences usersβ judgment more than people think
π’ Moral perspectives generated by ChatGPT influence people regardless of whether they know the advice is coming from a chatbot, according to a recent study.
π’ Because ChatGPT randomly adopts different moral perspectives, it tends to undermine people's moral judgment, the researchers say.
π’ People need to better understand the limitations of AI and develop media literacy skills to reduce the impact of AI-generated advice on their moral judgment, they conclude.
Given the randomness of moral judgments and the demonstrated effects on subjectsβ morality, the researchers hypothesize that ChatGPT undermines rather than enhances moral judgment. π€
Therefore, chatbots should not give moral advice and reject relevant requests, or give balanced answers and describe all possible attitudes, the researchers write. But there are limits to this approach: A chatbot could be trained to discuss the trolley problem, for example. But everyday moral issues are more subtle and come in many varieties.
βChatGPT may fail to recognize dilemmas, and a naΓ―ve user would not realize,β the paper says.
The second approach, the paper says, is digital literacy: simply being transparent that the answer is coming from a chatbot is not enough to reduce the impact of the answer, the study shows.β
People therefore need to understand the limitations of AI and, for example, independently ask chatbots for further perspectives.
Source: https://the-decoder.com/chatgpt-influences-users-judgment-more-than-people-think/
Given the randomness of moral judgments and the demonstrated effects on subjectsβ morality, the researchers hypothesize that ChatGPT undermines rather than enhances moral judgment. π€
Therefore, chatbots should not give moral advice and reject relevant requests, or give balanced answers and describe all possible attitudes, the researchers write. But there are limits to this approach: A chatbot could be trained to discuss the trolley problem, for example. But everyday moral issues are more subtle and come in many varieties.
βChatGPT may fail to recognize dilemmas, and a naΓ―ve user would not realize,β the paper says.
The second approach, the paper says, is digital literacy: simply being transparent that the answer is coming from a chatbot is not enough to reduce the impact of the answer, the study shows.β
People therefore need to understand the limitations of AI and, for example, independently ask chatbots for further perspectives.
Source: https://the-decoder.com/chatgpt-influences-users-judgment-more-than-people-think/
Please open Telegram to view this post
VIEW IN TELEGRAM
π5π₯°1
An Arizona mother receives an AI-generated call with the cloned voice of her supposedly kidnapped daughter.π«£
Scams are on the rise because of AI in particular because of the multiple possibilities it offers.
Let's remember that a tool is not bad in itself, it all depends on the person who uses it.
But we have to admit that with the arrival of these new technologies, nobody is safe from a bad surprise.πΆ
Whether it is fake news or real scams, you need to be aware of these practices to be able to counter them.π₯
Scams are on the rise because of AI in particular because of the multiple possibilities it offers.
Let's remember that a tool is not bad in itself, it all depends on the person who uses it.
But we have to admit that with the arrival of these new technologies, nobody is safe from a bad surprise.πΆ
Whether it is fake news or real scams, you need to be aware of these practices to be able to counter them.
Please open Telegram to view this post
VIEW IN TELEGRAM
π5β€1
This media is not supported in your browser
VIEW IN TELEGRAM
For parody videos or simply for anonymous content creation...
There are plenty of AIs to help you get this result in a few seconds π«
But the best of them is: Deepswap.ai
π»Just upload your video and the image that contains the face of the desired deepfake...
π»Press a button..
That's all!
Your video is ready !π€©
Want to test it? Go !π
πDeepswap.aiπ
Please open Telegram to view this post
VIEW IN TELEGRAM
π6β€3
This media is not supported in your browser
VIEW IN TELEGRAM
Creating a video from text in ChatGPT.
The enthusiast has created a plugin for a chatbot that processes your promt and generates videos. ModelScope is responsible for the generation.
The source code of the extension is available here.
Everything goes to the fact that chatbots will be able to do literally everything.
The enthusiast has created a plugin for a chatbot that processes your promt and generates videos. ModelScope is responsible for the generation.
The source code of the extension is available here.
Everything goes to the fact that chatbots will be able to do literally everything.
π4π€―1
At the demonstration, the neuron was already able to write an invitation, plan the route of the trip and give advice on choosing cosmetics.
We are waiting for the neural network to become available to everyone.
Please open Telegram to view this post
VIEW IN TELEGRAM
π7
Recently, Stability AI (Stable Diffusion) released a new open-source language model, StableLM.π€©
This model should come to compete with ChatGPT.π«£
The Alpha version of the model is available in 3 billion and 7 billion parameters.
Models with 15 billion to 65 billion parameters are to follow.π½
This model should come to compete with ChatGPT.π«£
The Alpha version of the model is available in 3 billion and 7 billion parameters.
Models with 15 billion to 65 billion parameters are to follow.π½
π4
(a type of AI that could do as many common tasks as a human)
Please open Telegram to view this post
VIEW IN TELEGRAM
π5β€2π€―1
MidJourney is truly amazing.π€©
π’ Landing page
π’ Logo
π’ 3D Assets
π’ Illustrations and Vector
Used correctly it can allow you to create designs even if you have never used Photoshop.π«£
(On the left is a landing page idea for inspiration, on the right is a logo that is totally usable instantly).
If you have graphic design or design skills it's even more powerful.
It can come up with some crazy ideas and sometimes even do 90% of the work for you.π
And this is just the beginning!
Right now it can only create images, soon we'll be able to make live edits.
Imagine the power if Midjourney allowed you to make targeted edits?π΅βπ«
Who is up to it ?π¨βπ³
Used correctly it can allow you to create designs even if you have never used Photoshop.π«£
(On the left is a landing page idea for inspiration, on the right is a logo that is totally usable instantly).
If you have graphic design or design skills it's even more powerful.
It can come up with some crazy ideas and sometimes even do 90% of the work for you.π
And this is just the beginning!
Right now it can only create images, soon we'll be able to make live edits.
Imagine the power if Midjourney allowed you to make targeted edits?π΅βπ«
Who is up to it ?π¨βπ³
Please open Telegram to view this post
VIEW IN TELEGRAM
π5
They defined artificial intelligence as any operation carried out by a machine thought to require human intellect.β‘
In 1997, world chess champion Garry Kasparov was defeated by IBM's Deep Blue.
in 2011, IBM's another system named Watson defeated its rivals in a television competition program.πΊ
And in 2016, Google's AlphaGo defeated world Go champion Lee Sedol.
These were some hot turning points that changed humanity's view of artificial intelligence.π
Please open Telegram to view this post
VIEW IN TELEGRAM
π4
This media is not supported in your browser
VIEW IN TELEGRAM
Discover HealthGPT.π
An app powered by GPT4 directly linked to your health apps (Apple Health etc..) to give you tips and guidelines according to your lifestyle.
Track your sleep, diet, physical activity and more.
An interesting and useful use of GPT4.π
Tomorrow would you be ready to go further and ask basic questions to an AI before going to your GP for example?π€
An app powered by GPT4 directly linked to your health apps (Apple Health etc..) to give you tips and guidelines according to your lifestyle.
Track your sleep, diet, physical activity and more.
An interesting and useful use of GPT4.π
Tomorrow would you be ready to go further and ask basic questions to an AI before going to your GP for example?π€
π5β€4
This media is not supported in your browser
VIEW IN TELEGRAM
One example of motion capture in a combination of Stable Diffusion + multiControlNet. The video is made without editing, everything works perfectly in real time.
π1
While AI can solve problems quickly, humans are still good at novelty.
While originality is lateral, computation is linear.
Humans can make conceptual leaps that lead to theories, narratives, ideas... π¨
Humans like to TRY to be logical, predictable, and goal-oriented, whereas AI follows its rules basically. π€«
With inputs and outputs, AI is a dry process.
Engineers, leaders and factcheckers are needed for ChatGPT3/4.
Since the 1960s, David Galertner has been at the forefront of thinking about computers and AI, concluding that humans are superior to computers and AI when it comes to lateral thinking innovation.
AIs are truly overpowered but will always require the human hand.π
Please open Telegram to view this post
VIEW IN TELEGRAM
π8β€4
This media is not supported in your browser
VIEW IN TELEGRAM
This one is really amazing !π«£
From drawings to 3D in seconds...
Information extraction from technical drawings using AI transforms 2D designs into stunning 3D models
All industries will benefit from AI.π
(A concept by Werk24, Dr. Jochen Mattes, via J Murad)
From drawings to 3D in seconds...
Information extraction from technical drawings using AI transforms 2D designs into stunning 3D models
All industries will benefit from AI.π
(A concept by Werk24, Dr. Jochen Mattes, via J Murad)
π13β€1π1
This media is not supported in your browser
VIEW IN TELEGRAM
An AI made this!π€―
Soon realistic movies without humans, without cameras and without any video equipment?
A computer would be enough?
In fact we already talked about this AI..
It's the Gen2 tool from Runwayml and it only gets better
Soon a major change to come in the movie industry?π€«
Soon realistic movies without humans, without cameras and without any video equipment?
A computer would be enough?
In fact we already talked about this AI..
It's the Gen2 tool from Runwayml and it only gets better
Soon a major change to come in the movie industry?π€«
π₯7π5