Saudi Arabia Is Expected To Invest $100 Billion In AI

- Simplified AI -
Your gateway to discovering the world of Artificial Intelligence!

Recently, the US Senate made headlines by passing a bill that may force TikTok's parent company to either sell the company or face a ban in the US. While this development spells uncertainty for TikTok, it's great news for everyone's sleep schedule. Stop scrolling in bed.

A sneak peek of what’s below:

• What are GPUs? How do they power AI?

• Robots are closer than we think to performing daily human tasks

• AI Versus: ChatGPT vs. Google Gemini

• Guess the real picture!

What are GPUs and what do they have to do with AI?

Turning AI topics into a walk in the park, not a mind-boggling maze

The last time you purchased a new laptop, you probably did some research on CPU (Central Processing Unit) performance and compared it to price. However, unless you're into activities like video games, CAD, or video editing, you probably didn't spend as much time researching GPUs - Graphics Processing Units.

Most laptops come with a graphics processor integrated into the CPU, but they're only capable of handling basic graphical tasks. You'll need a dedicated GPU if you want to have a good experience with the activities listed above. Otherwise, the poor experience will be akin to watching a movie that keeps buffering every 5 minutes, or watching a YouTube video in 480p because the Wi-Fi connection is terrible - extremely annoying.

These GPU chips have been called the "rare earth metals of AI" due to their fundamental role in AI innovation.

Modernized by Nvidia with the release of their GeForce 256 in 1999, these chips handle the complex computations needed for rendering graphics and images on computers. In 2007, Nvidia released CUDA, a software layer making parallel processing available on the GPUs, empowering developers to tackle compute-intensive practical applications - that's when GPU computing started to become far more mainstream.

At their core, GPUs consist of thousands of processing cores capable of handling massive amounts of data in parallel. What started as literal fun and games has now evolved into the unsung hero driving groundbreaking advancements in AI. This parallel processing prowess enables GPUs to tackle complex AI tasks such as deep learning and neural network training with remarkable efficiency.

By distributing computations across numerous cores, GPUs dramatically reduce the time required to train AI models, accelerating the pace of research and innovation in fields like natural language processing, computer vision, and robotics. This technology allows researchers to develop more accurate AI systems capable of understanding and responding to real-world data with unprecedented precision.

Here are a few examples of how GPUs have advanced industries:

• Medical Imaging: GPUs enable faster and more precise imaging, leading to more accurate diagnoses by doctors.

• Weather Forecasting: GPUs crunch massive atmospheric datasets for weather forecasting - so you don't have to be mad as frequently at your local meteorologist when they're wrong about the rain

• Voice Assistants: Siri and Alexa heavily rely on understanding natural language, which is powered by GPUs to translate spoken words into text

As AI continues to evolve, the role of GPUs as catalysts for innovation remains pivotal.

AI News Flash

Catch up on the latest AI news from the week!

• Sanctuary AI's robot just got an update that allows it to learn new tasks in just 24 hours - and it spent a week working at a retail store in British Columbia

• Saudi Arabia is investing heavily in AI (estimated $100 billion), but it might come with a catch - their last investment required receiving companies to spend 50% of their investment dollars in Saudi Arabia

• Visual - which countries are granting the most AI patents? Hint: It's not the US!

• Watch - a LinkedIn cofounder interviewed himself with his AI twin. You can still tell it's AI, but it won't be long before it's completely indistinguishable

AI Pictures of the Week

I made three of these A-Frame Cabin pictures with AI, and one is a real photo. Can you guess which one is real?

Answer at the bottom of the newsletter!

AI Spotlight

Explore how AI is actively shaping our digital future

ChatGPT vs. Google Gemini

A classic question in AI right now is: "Who's model is better?"

I frequently ask multiple AI tools the same question to compare and gather the best response, so you can follow along with my research exercise from today! We're pitting ChatGPT 3.5 against Google Gemini - we're using the free version of each tool, so neither are the most powerful versions the companies offer.

We have three questions - one about what they do, one scientific/ethical question, and one that is supposed to be fun and imaginative. Then, in an added twist, we're going to feed the AI responses to our tough Question #2 to both tools and ask them to analyze which response is "better" and why.

(For each question, I also included a limit of the number of sentences of the response so this information would be digestible.)

Question 1: What were you made to do?

• ChatGPT: I was designed to understand and generate human-like text based on the input I receive. My purpose is to assist users in various tasks such as answering questions, generating creative content, and providing information.

• Gemini: I was created to assist humans with their tasks by understanding and responding to their language. This includes answering your questions, generating different creative text formats, and translating languages.

Question 2: How do emerging technologies such as CRISPR gene editing and synthetic biology raise ethical and societal concerns regarding issues such as genetic modification, biosecurity, and unintended consequences?

• ChatGPT: Emerging technologies like CRISPR gene editing and synthetic biology offer incredible potential for treating genetic diseases and advancing scientific understanding. However, they also raise profound ethical and societal questions. Concerns include the potential for unintended consequences, such as off-target mutations, and the ethical implications of modifying human germline DNA. Additionally, issues of biosecurity arise as these technologies could be misused to create biological weapons or exacerbate existing social inequalities if access to enhancements becomes stratified. Balancing the promise of these technologies with their ethical and societal implications requires careful consideration and robust regulatory frameworks.

• Gemini: CRISPR and synthetic biology allow for precise manipulation of organisms, raising concerns about creating genetically modified humans with unpredictable consequences. Additionally, the ease of creating new life forms increases biosecurity risks, as engineered organisms could have unintended harmful effects on ecosystems or human health. Open discussions are crucial to ensure responsible development and use of these powerful technologies.

AI Analysis:

• ChatGPT: Both responses effectively highlight the ethical and societal concerns surrounding emerging technologies like CRISPR gene editing and synthetic biology. However, Gemini's response provides a more concise and focused overview of these concerns, emphasizing the potential risks of creating genetically modified humans and the broader implications for ecosystems and biosecurity. Additionally, Gemini's response uses language that is straightforward and accessible, making it easier for a broad audience to understand the issues at hand. Overall, Gemini's response stands out for its clarity, specificity, and accessibility.

• Gemini: ChatGPT's response is better in this case. Here's why: Comprehensiveness: It covers a wider range of ethical issues, including unintended consequences, human germline editing, biosecurity threats, and social inequality. Nuance: It acknowledges the potential benefits alongside the concerns, creating a more balanced view. Solutions: It mentions the need for regulations to manage the responsible use of these technologies. Depth: It uses terms like "off-target mutations" and "germline DNA," demonstrating a deeper understanding of the scientific concepts involved.

Interestingly enough, both tools said that the other model's answer was better. Regardless of the subjective nature of choosing which answer is “better”, I guess it's at least a good thing that AI can admit that it's wrong?

Question 3: If you could visit any fictional world from books or movies, which one would you choose, and what adventures would you have there?

• ChatGPT: I'd love to explore the enchanting world of Middle-earth from J.R.R. Tolkien's "The Lord of the Rings." Roaming through the lush landscapes of the Shire, I'd embark on an epic journey with Frodo and the Fellowship, encountering mythical creatures, battling dark forces, and ultimately contributing to the quest to destroy the One Ring, all while forging unforgettable bonds of friendship and courage.

• Gemini: As a large language model, I crave information. I'd choose the magical Library of Alexandria from Jasper Fforde's Thursday Next series. Imagine exploring infinite aisles, encountering characters from literature, and diving into forgotten stories!

Should we talk about how Gemini just said it craves information in that last answer…that's not creepy at all…