New York City Public Schools Ban OpenAI's ChatGPT

Welcome to Daily Zaps. Think of us as your personal AI wingman, here to keep you up-to-date and entertained. So sit back, relax, and let us do the heavy lifting. It's going to be a wild ride in 2023!

Here's what we got for you today:

• New York City Public Schools Don't Love ChatGPT 💔

• More Microsoft + OpenAI News 🤖

• ChatGPT Is Helping People with ADHD ✅

• Cool Ai Apps and News 💻

• Ai Meme of the Day ⚡

New York City Public Schools Ban OpenAI's ChatGPT

Well, it looks like New York City public schools are really not taking any chances when it comes to cheating! They've banned ChatGPT, an AI chatbot developed by OpenAI, from school devices and WiFi networks. And can you blame them? ChatGPT, or "generative pre-trained transformer" (sounds like something out of a sci-fi movie), is like a text-producing version of AI art. It can answer fact-based questions and write essays and articles that are often indistinguishable from human-written content. I mean, talk about unfair competition! It's like these students are trying to take over the world or something...

But seriously, while ChatGPT may be able to provide quick and easy answers, it doesn't necessarily help students develop critical-thinking and problem-solving skills, which are essential for academic and lifelong success. So, I can see why the schools are worried about it being used for cheating. But, let's be real, it's not like students won't find a way to use it on their own devices with cellular networks or non-school WiFi. And, as AI improves, it will only become harder to tell the difference between human-written and AI-generated content.

However, not everyone in the education community is against ChatGPT. Some teachers argue that it could actually be an ally, allowing them to use it as a baseline essay response that the class can then work together to improve upon. The key, according to these teachers, is to focus on encouraging students to "explore things worth knowing" rather than just pursuing standardized metrics.

At the end of the day, whether schools choose to embrace or ban AI like ChatGPT, it's clear that it's here to stay. And with that comes both the potential for incredible advancements, as well as the potential for societal fallout, including the automation of more and more jobs. It's going to be interesting to see how all of this plays out, and whether or not we'll see government regulation in the near future (though, with the current trajectory of the US Congress, it doesn't seem likely). In the meantime, we'll just have to wait and see how ChatGPT and other AI developments impact the education system and the world at large.

Microsoft Office + OpenAI On Steroids

Microsoft is planning to bring some seriously smart AI chatbot technology from OpenAI into their Office tools and apps? Microsoft has been thinking about adding the AI technology to Word, PowerPoint, Outlook and other apps to generate text with just a few prompts. Crazy, right? And get this - the chatbot will also be added to Bing, so you can create emails and other texts with it too. Plus, Microsoft is planning to integrate image-generation software from Dall-E 2 into Bing, so you can search for images based on written descriptions instead of just keywords. That's some next level stuff right there.

ChatGPT Is Helping People with ADHD

A Reddit poster has found that using a language model like Assistant (called ChatGPT in this case) has helped them improve their performance in computer science and do better on exams, especially in understanding difficult concepts. They prefer using ChatGPT to traditional resources like Google, YouTube, and textbooks because it allows them to have a conversation and ask for clarification, and helps them understand dense material more efficiently. [Link]

Cool Ai Apps + News

AI Internship at Microsoft: Microsoft is currently looking for research interns to help develop the next generation of productivity apps using large language models. They're especially interested in candidates with expertise in areas like augmentation of language models, prompt engineering, hallucination mitigation, and other applied research subjects related to language model interaction and efficiency. If you're skilled in these areas and want to join a dynamic team of top scientists and engineers, consider applying for a research internship at Microsoft! [Link]

VALL-E: The Microsoft folks are really going all in on AI and they developed this really cool method called VALL-E that can turn text into speech. It works by using these special codes that come from another process that turns sound into numbers. This is different from other methods that try to match the sound to a set of numbers. They used a ton of data to train VALL-E - like 60,000 hours of people talking. That's hundreds of times more data than other methods have used. And it worked really well because VALL-E can make speech that sounds super natural and like the right person, even if it's only heard them for a few seconds. Plus, it can keep the emotions and the sounds of the place where the person was talking when it makes the speech. Pretty cool, right? [Link]

GPTZero: GPTZero is an app that was developed by Edward Tian, a student at Princeton University, to help fight against the increasing issue of AI plagiarism. According to what Tian said on Twitter, GPTZero is able to quickly and accurately determine if an essay was written by ChatGPT or a human. I decided to test is out using ChatGPT results for "write a 5 paragraph essay about why christopher columbus did not discover america" [Link]

GPTZero analyzes the complexity and repetition of the content to determine if it's computer-generated or not. Complexity, or perplexity, is how well a computer model can guess the next word in a sentence. The lower the complexity, the better the model is at predicting. Repetition, or burstiness, is how often certain words or phrases are used in a short time. By looking at both of these factors, we can tell if the text was written by a computer or a person. GPTZero provides a lot of technical details about how it analyzes the data, and then gives you a button to see the final conclusion.

🚨Beware students and adults who may be thinking of trying to trick their teachers and professors! The app correctly identified that my content was generated by AI. Be cautious when using this ChatGPT or similar apps to get help with homework. Use AI as a helper not a full generator of your work. 😳

Konjer: Have you ever been like, "Ugh, why do I gotta read this whole book just to find the answer to one question?" Same, my dude. But fear not! Konjer has got your back with their AI-powered web app that lets you holla at books and get the 411 you need without all that pesky reading. And hey, robots are rad as hell, amirite?" [Link]

Peace out,Daily Zap Team

Ai Meme of the Day 😂