AI roundup👩‍💻

AI roundup👩‍💻

👦 ChatGPT takes over South Park

You can watch the full trailer here.

The boys of South Park are leveraging machine learning for dating purposes! In the above clip, Stan asks Clyde for dating advice, noting that he always seems to say the wrong thing to his girlfriend. Clyde responds by saying he uses ChatGPT language AI to generate text replies to his girlfriend. Ooooh boy.

The episode will examine OpenAI's ChatGPT combined with its own perspective to spark a stimulating discussion. As always, fans should be prepared for the unexpected and anticipate the show to take a controversial and audacious stance on the growth of artificial intelligence.

Interestingly though, we see the real twist at the end of this episode...

Sources: WindowsCentral, IMDb

🤖 Say hello to Discord's advanced AI chatbot

Discord, a popular platform that lets users chat by text, video, and voice, will soon enable artificial intelligence (AI)-based features powered by OpenAI technology. Discord wrote that the company is excited about “the opportunities for AI to enhance so many facets of human experience.”

Clyde: This AI-powered chatbot can answer questions and have conversations. You can type "@Clyde" in a server to talk to the chatbot in any channel. Clyde can also be used to start a thread for a group of people to hang out with. The chatbot can recommend playlists, and access GIFs and emojis like any Discord user.

CEO Jason Citron explained Discord provides, "AI at your fingertips, together." Discord aims to provide three AI experiences as free public experiments in a limited number of servers: an AI chatbot, an automated AI moderator, and a conversation summarizer.

Sources: Livemint, TechRadar

🚀 Revolutionizing AI with GPT-4

Microsoft Store sign.
Photo by Turag Photography / Unsplash

On March 9, 2023, Microsoft Germany CTO Andreas Braun announced the imminent release of GPT-4 at an AI kickoff event entitled "AI in Focus - Digital Kickoff."

GPT-4 is 500 times more powerful than the current ChatGPT. The current version of ChatGPT has about 175 Billion Machine Learning Parameters (ML parameters). The new GPT-4 will have over 100 trillion ML parameters. This will allow GPT-4 to process different types of data — videos, images, sounds, numbers, and many more.

Sources: LinkedIn, EconomicTimes, YugaTech

📸 Visual ChatGPT: More than just text

Print is Alive
Photo by Bank Phrom / Unsplash

A system that includes multiple visual foundation models and allows users to communicate with ChatGPT via graphical user interfaces. This combines various visual foundation models, such as Transformers, ControlNet, and Stable Diffusion, with ChatGPT. With the use of these visual foundation models, ChatGPT can now handle user requests for image generation and editing.

Visual ChatGPT is capable of handling complex visual queries or editing instructions that require the cooperation of multiple AI models across several phases. Researchers concluded that a self-correcting module is necessary to ensure execution outputs align with human intentions and to make necessary adjustments.

What can this do?

Sources: EconomicTimes, Livemint, Github

🏥 The use of AI for faster medical procedures

This image was captured in 1958, and depicted a physician and nurse, who were examining a tuberculosis positive chest x-ray.
Photo by CDC / Unsplash
"Chest X-rays are one of the most common imaging examinations performed worldwide. Even a small percentage of automatization can lead to saved time for radiologists, which they can prioritize on more complex matters."
- Dr. Louis Lind Plesner

Researchers used a commercially available AI tool to analyze the chest X-rays of 1,529 patients from four hospitals in the capital region of Denmark. Chest X-rays were included from emergency department patients, in-hospital patients, and outpatients. The X-rays were classified by the AI tool as either "high-confidence normal" or "not high-confidence normal" as in normal and abnormal, respectively.

Sources: ScienceDaily

👩‍🚀 The role of artificial intelligence in Mars Exploration

Photo by Christian Lischka SJ / Unsplash

A team of astrobiologists developed an AI model and tested its ability to look out for sparse life hidden away in salt domes, rocks, and crystals at Salar de Pajonales at the boundary of the Chilean Atacama Desert and Altiplano - one of the driest places on the planet, resembling the features of a Martian surface.

The AI algorithm consists of Convolutional Neural Networks (CNNs), which were trained using the data collected to predict large-scale geological features and micro-habitats that have the highest likelihood of having biosignatures. "While the high rate of biosignature detection is a central result of this study, no less important is that it successfully integrated datasets at vastly different resolutions from orbit to the ground, and finally tied regional orbital data with microbial habitats," said Nathalie A. Cabrol from the SETI Institute NAI team.

“Artificial intelligence-machine learning models successfully identified geologic features with high probabilities for containing biosignatures at spatial scales relevant to rover-based astrobiology exploration,” the scientists concluded.

Sources: PsychologyToday, Interesting Engineering

That's all for this week folks! Check out our other blogs.