I am quitting ChatGPT

Welcome to Daily Zaps. Imagine us as your AI party planner, here to make sure you have a good time. So kick off your shoes, let down your hair, and let us handle the details. It's going to be a wild ride in 2023!

Here's what we got for you today:

• Are people addicted to ChatGPT? 💉

• OpenAI is working on a watermarking tool to mark GPT generated content 🔬

• Google is planning to release their own chatbot, called Sparrow 🧠

• People Tracking Using WiFi + AI 📶

Addicted to AI Technology Like ChatGPT?

Below is a screen shot from a reddit post where someone is finding that ChatGPT was too addicting and causing problems for them. In response to post someone replied with a name for potential disorder and definition - AIDD, or AI Dependence Disorder. AIDD, occurs when a user of AI offloads a great deal of cognitive burden onto AI software and the service later becomes inaccessible, causing a sort of digital withdrawal. This can result in feelings of claustrophobia, loss of agency, depression, and helplessness as the user realizes they will now need to again devote the time and energy they previously freed up through an AI assisted workflow, to what now seem like menial tasks. For those suffering from AIDD, the AI comes to feel essential, in an existential sense, to maintaining space and freedom. Without the AI, a seemingly crushing cognitive burden is again lowered onto their shoulders, where previously that time could have been devoted to rest, leisure, and personal development. [Link]

OpenAI Watermarking AI / GPT Generated Content  

OpenAI is working on this tool that's a bit like a secret code for text generated by AI like GPT. The idea is that when GPT spits out a bunch of text, there's a hidden signal in there that can prove it came from GPT. This makes it harder for people to pass off AI-generated text as human-written, which can be helpful in preventing plagiarism and stopping trolls from impersonating people online. It's like a watermark for AI text, if you will. The tool uses a special code that only we know, to add this signal to the text, so it's not noticeable to the end user but can be detected later. Cool, huh? [Link]

DeepMind / Google Sparrow - ChatGPT Competitor

So the DeepMind squad is thinking about putting out their own chatbot, Sparrow, for a little test run next year. They wanna take their sweet time and add some cool new features like being able to cite sources, which ChatGPT can't do yet. They're gonna be using reinforcement learning to make it happen.[Link]

People Tracking Using WiFi + AI

Scientists have come up with a new way to track people's full body movements using just WiFi signals. They created a deep learning algorithm that can understand the different signals from WiFi and map it to different body parts. Basically, it can track multiple people just by using WiFi signals, pretty cool right? And the best part is, WiFi is almost everywhere and can be added to any environment very easily, this opens up so many possibilities for both personal and business use cases. [Link]

ChatGPT Prompt Hack

When you're using something like ChatGPT, it's kind of like using Google search. You type in what you're looking for, but the key to getting really good answers is to give ChatGPT really specific instructions on what you want. These instructions are called prompts. Think of it like asking a librarian for a book, you want to give them as much information as possible so they can find the exact book you're looking for. The same goes for ChatGPT, the more specific you are with your prompts, the better the results will be. So, before you start asking ChatGPT questions, take a moment to think about exactly what you want to know and how you can phrase it in a prompt. Trust me, it'll save you a lot of time and give you much better answers. [Link]

Peace out,Daily Zap Team