- Simplified AI
- Posts
- ChatGPT goes to college!
ChatGPT goes to college!
Plus: What is Deep Learning? And Rabbits aren't just good for magic tricks...
- Simplified AI -
Your gateway to discovering the world of Artificial Intelligence!
Ok summer, I judged you too harshly. With low temps and heavy wind chills making it feel like 2° this week, I'm ready to get back to warmer weather and start wearing shorts again. I still haven’t hit my limit of “it’s too cold out to eat ice cream” though - that’s probably somewhere around -10°.
As I dream of warm, sunny days, let's dive into some exciting AI topics!
A sneak peek of what’s below:
What is deep learning for AI?
The first major university announced they're bringing AI on campus
A new handheld AI device is promising a new way for you to interact with your apps
Get creative: Create an image of the Northern Lights!
Was this newsletter forwarded to you? Subscribe to get Simplified AI delivered straight to your inbox every Sunday!
What is deep learning?
Turning AI topics into a walk in the park, not a mind-boggling maze
What is deep learning? Well it isn't jumping into the deep end of a pool and reading a book underwater if that's what you were thinking. Instead, shockingly, it has to do with AI. I know -who would have thought?!
It's a subset of artificial intelligence that mimics how the human brain functions in order to process data, recognize patterns, and make accurate predictions. It can be categorized into four major areas: computer vision, speech recognition, natural language processing, and recommendation engines.
The primary focus of deep learning is to reduce the manual labor involved in training AI models. Deep learning deals with "supervised" training, which we wrote about in a previous newsletter. However, instead of a human painstakingly labeling countless images or data points and then feeding it into an AI model to train it, deep learning algorithms autonomously identify and extract relevant features from data.
Take the example below - just as a person would use visual cues, textures and flavors to identify an unknown meal like below, deep learning models follow a similar process to arrive at accurate conclusions.
1) Ingredient Recognition: Your journey begins by examining the ingredients. You see long, delicate strands of pasta and immediately you recognize that it's a pasta dish and not a steak, drawing from your wealth of pasta experiences. However, you're still unsure of what exact type of pasta dish this is, so you need to go to the next layer - texture.
2) Texture Assessment: You notice al dente spaghetti coated with a creamy neutral-colored sauce. The absence of a red or green sauce begins to dismiss options like marinara and pesto. You feel pieces of crispy pancetta, and specks of freshly ground, crunchy black pepper. These textural clues paint a picture that this is spaghetti carbonara, but the connoisseur in you still needs more information to confirm, so you dig in and go to the next layer - flavor profiling.
3) Flavor Profiling: You taste the dish and a symphony unfolds within your palate as you encounter the creamy, slightly salty embrace of eggs and cheese in the sauce, harmoniously intertwined with savory pancetta and black pepper. It's a sensory symposium that leaves little doubt - you're enjoying a plate of spaghetti carbonara.
4) Confirmation: With a triumphant sense of assurance, you stand firmly by your conclusion. It is indeed spaghetti carbonara and not another type of pasta. You arrived at this conclusion by going through pattern recognition and various layers of analysis.
Why is this important?
Deep learning underpins numerous AI applications embedded in everyday products. From digital assistants like Siri and Alexa to voice-activated TV remotes, fraud detection systems, facial recognition technology, and even self-driving cars' ability to discern road signs and pedestrians, deep learning continues to reshape the way we interact with technology and automate complex tasks.
AI News Flash
Catch the latest AI news that is making all of the headlines this week!
AI In the Classroom
Arizona State University (ASU) is making waves in the world of education by joining forces with OpenAI. ASU's partnership with OpenAI grants the university access to the premium, Enterprise version of ChatGPT, propelling the institution to the forefront of innovative teaching and learning practices.
OpenAI and ASU's vision extends beyond the ordinary; it's all about enhancing the educational journey for both educators and students. ASU wants to leverage this technology to develop personalized AI tutors and to enhance research efforts across campus, among other things. Additionally, ChatGPT will revolutionize ASU's courses, aligning educational offerings with the dynamic demands of the ever-evolving tech industry.
The university is optimistic about the transformative potential of this collaboration. Large employers, like KPMG, have noticed a gap in recent graduates comfortability levels in using AI tools because they have been considered cheating tools in education settings since their launch. While this is expected, given the novelty of the tools, ASU is setting an example for other educational institutions grappling with the challenge of integrating AI into their curricula.
This partnership highlights the broader need for academia to keep pace with advancing technology. Students aspire to enter the job market feeling well-prepared, and AI proficiency is quickly going to become an expectation. ASU's groundbreaking collaboration with OpenAI represents a significant step toward addressing these challenges and preparing students for a future where AI is a fundamental part of their careers and daily lives.
Other Stories We're Following:
The World Economic Forum at Davos took place last week and AI was at the center of discussion
Meta (Facebook's parent company) is ramping up AI efforts after their Metaverse project failed
AI Company Spotlight
Explore trailblazing companies that are shaping our digital future
Rabbit
There's a new AI hardware device hopping onto the market called the Rabbit R1. (See what I did there?)
At first glance, the Rabbit R1 may conjure nostalgia for the handheld video game devices of the late 90s and early 2000s. However, this AI-powered marvel isn't about reliving the past; it's about shaping the future of app interaction. It offers users a unique way to manage and navigate a digital world with AI.
Think of it like an advanced Roku or Amazon Fire stick. It will serve as an aggregator for apps, allowing you to command each app within a singular device with voice control. Just like when you speak into your Fire Stick and say, "play 'Reacher' on Amazon Prime," and the system loads Amazon Prime Video and starts playing, the Rabbit R1 aims to take this functionality to a whole new level, offering a more extensive range of capabilities.
The device promises a lot and we'll have to wait and see what it delivers. The most interesting component to me is the Rabbit Hole online portal that connects to the device. This portal serves as the gateway not only for logging into all your apps but also for training the AI—a task that may be unfamiliar to many. The example the company provides uses Photoshop - they say you'll be able to log into the Rabbit Hole, start a screen recording session, and demonstrate to the AI how tasks like adding/removing watermarks from photos are accomplished. Once shared, the device should have the ability to automate that step for you when you instruct it to do so in the future. In theory, this could streamline various processes in both your work and daily life, all through simple voice commands to the AI.
There's a promise of an all-day battery life, but we all know how this story goes - battery never lasts as long as you want it to and it heavily varies on the usage of the user. Notably, the company decided to build their own operating software and AI "Large Action Model" and skip the use of ChatGPT or another big name model.
Unlike Humane's AI Pin that we covered in a previous newsletter, Rabbit's CEO has been clear that their device is not meant to replace your smartphone. But let's be very clear that smartphones are direct competition for devices like these. These types of products will have to REALLY wow people to get them to move away from being glued to their smartphones, and I think it's unlikely that the first few AI hardware products we'll see will make any significant ground in that, but it's a first step in getting people comfortable with a different type of device and seeing what they're capable of.
Was this newsletter forwarded to you? Subscribe to get Simplified AI delivered straight to your inbox every Sunday!
Get Creative with AI
Practice using AI and unleash creativity with your new collaborator
Copy the prompts below and paste them into ChatGPT, BingAI, Google Bard, or your generative AI tool of choice and let AI work its magic. Share your results with us…and take it a step beyond the initial prompts and see what you can create!
Text-based
You wake up one Saturday morning and find that you've switched bodies with your dog. Describe the hilarious adventures and challenges that ensue throughout the day until you switch back the following morning.
Visual-based*
Create an image of a starry Arctic night with the Northern Lights dancing across the sky. Two individuals, bundled up in warm winter attire, are standing in awe and wonder as they gaze at the Aurora Borealis.
*Visual-based examples are for users of Dall-E, BingAI, or premium users of ChatGPT - may require paid subscription.