How to use custom GPTs inside ChatGPT

OpenAI's new scale for AGI, Amazon's shopping assistant goes live, a guide to custom GPTs and more!

Author

Everyday AI
July 12, 2024

Outsmart The Future

Today in Everyday AI
7 minute read

🎙 Daily Podcast Episode: Not sure what GPTs are? Wondering if you should make one inside ChatGPT? We explain how they work and why you should have one. Give it a listen.

🕵️‍♂️ Fresh Finds: Why the Atlantic partnered with OpenAI, Microsoft adds Copilot to OneNote, a new ElevenLabs competitor teases update and SoftBank acquires an AI chipmaker. Read on for Fresh Finds.

🗞 Byte Sized Daily AI News: OpenAI unveils its progress scale for AGI, Amazon’s shopping assistant Rufus goes live and EU’s AI Act finally set in place. For that and more, read on for Byte Sized News.

🚀 AI In 5: We’re giving today’s episode the AI in 5 treatment and breaking down why you should build a custom GPT. See it here

🧠 Learn & Leveraging AI: If you’re looking to understand what GPTs are or how to build one, we have the perfect guide for you. Keep reading for that!

↩️ Don’t miss out: Did you miss our last newsletter? We talked about Google powering a robot with Gemini, AWS making big AI moves and AI's impact on the environment. Check it here!

How To Use Custom GPTs Inside ChatGPT 🤖

What the heck is a GPT and do I need one?

The answer is yes.

GPTs take ChatGPT to the next level, allowing you to create custom workflows for your specific needs.

Whether you're a beginner or you've built a GPT or 5, this episode is for you.

Join the conversation and ask Jordan questions on GPTs here.

Also on the pod today:

• Using GPTs in everyday tasks 🛠️
• Monetization of GPTs in GPT Store 💰
• Automating work operations with GPTs 🔁

It’ll be worth your 55 minutes:

Listen on our site:

Click to listen

Subscribe and listen on your favorite podcast platform

Listen on:

Here’s our favorite AI finds from across the web:

New AI Tool Spotlight – SuperGPT lets you use ChatGPT on Google Sheets, Telescope uses AI to find the right audience or employee, and Intrō is an AI-powered marketing automation for Instagram.

Trending in AI - The Atlantic’s CEO spoke on why it decided to make a deal with OpenAI.

Big Tech - Microsoft is adding Copilot AI to OneNote so it can read and analyze your handwritten notes.

AI Voice Tech — Murf AI, a text-to-voice AI platform and competitor to ElevenLabs, teased their new Gen 2 model.

Business of AI – SoftBank is acquiring UK AI chipmaker Graphcore.

AI Startups - A former Google employee has created an AI startup for drug development.

Read This - This study shows that most ChatGPT users think AI models have conscious experiences.

1. OpenAI's Scale for AGI Progress 🧠

In a bid to achieve Artificial General Intelligence (AGI), OpenAI has introduced a new grading system to track their advancements, with levels ranging from basic problem-solving to creating new innovations and performing work for entire organizations. The company's unique structure is centered around the mission of achieving AGI, with a commitment to assisting value-aligned and safety-conscious projects that aim to build AGI.

Despite varying timelines and uncertainties, OpenAI's CEO, Sam Altman, optimistically suggested AGI could be within reach in "five years give or take."

2. Amazon Launches AI Shopping Assistant "Rufus" in the U.S. 🛍

Amazon has introduced "Rufus," an AI-powered shopping assistant, located in the bottom right of the Amazon mobile app, aiming to enhance the shopping experience for customers in the US. Rufus is designed to assist users in finding products, making product comparisons, and receiving tailored recommendations based on their preferences.

Customers can ask Rufus a variety of questions about products, including factors to consider before purchasing, product reviews, and expert analysis sourced from around the web.

3. EU’s AI Act To Be In Effect By 2026 🇪🇺

The European Union is set to enforce new regulations on artificial intelligence applications by mid-2026, with different compliance deadlines for various AI use cases. From banning high-risk AI applications to imposing transparency requirements on developers, this landmark legislation aims to ensure responsible AI use.

With concerns raised by industry players and Member States, the EU is navigating a path towards fostering homegrown AI innovation while competing with global tech giants.

4. U.S. Senate Introduces COPIED Act to Protect Content Origin 🇺🇸

A bipartisan group of senators introduced the COPIED Act to safeguard journalists and artists from AI-generated content theft. The bill directs NIST to establish standards for content origin verification and synthetic content detection, ensuring creators' rights are protected.

With powerful sponsors like Senate Commerce Committee Chair Maria Cantwell, the bill aims to prevent unauthorized usage of creative works.

Our hot take — we’ve been saying for more than a year that there won’t be any meaningful legislation passed in the U.S. regarding AI and copyright.

5. Tech Giants' AI Race Spells Trouble for the Environment 🤖

Google and Microsoft's fierce competition in the AI realm is leading to a significant surge in energy consumption and CO2 emissions. Research reveals that AI advancements are driving the need for massive data centers, resulting in a substantial environmental impact.

With plans to achieve net-zero emissions by 2030, these tech giants are facing challenges as their AI projects require more electricity, contributing to the growing environmental crisis.

We talked about this EXACT issue yesterday in case you missed it.

Learn how to make a custom GPT!

Wondering what a Custom GPT is and how it can revolutionize your workflow?

Whether you're a seasoned tech pro or a curious newbie, this guide is your first step towards optimizing your knowledge work using AI.

We break down the essentials of custom GPTs in ChatGPT and how you can tailor this powerful tool to fit your specific needs.

Or see this related video:

🦾How You Can Leverage:

What in the actual heck is a GPT and do I need one? 

If you’ve seen the buzz over the past few months but don’t understand the hoopla, we’ve got you covered. 

Here’s the teaser — you should absolutely be building and deploying GPTs. 

Like…. All the time. 

And it’s a lot easier than you may think.

Today, we went elementary on GPTs, going over the most asked questions, common misconceptions, and a bit of myth-busting. 

If you’re wondering if GPTs are for you or your biz, we answer that and more in today’s show. 

Ready to get down to GPT bizness? 

Here’s the 1, 2, 3 of what ya need to know. 

👇

1 – Your skills aren’t the special 🤷

Not trying to be mean here y’all. We’re not special either.

Just keeping it real. 

(And wanting you to understand how knowledge work…. Works.) 

We assume that most people reading this are knowledge workers. In other words, a company pays you to sit in front of a computer, use your expertise in a certain domain or skillset, and create value for them. 

  • You read. 

  • Analyze. 

  • Prioritize. 

  • Create. 

  • Collaborate. 

  • Repeat. 

Guess what? 

If you know what you’re doing, Large Language Models can do almost all of those things better than us experts.

Hard pill to swallow. But pills are good! 

Think of it this way — how many times a day do you Google something? Or read something from a competitor? Or look up a stat from a white paper? 

Most knowledge workers spend the overwhelming majority of their time repeating manual tasks that AI now does better. 

Try this:

Don’t believe us? Need some high-priced consultancy to convince you otherwise? 

Check this McKinsey digital study that shows how up to 70% of tasks that employees spend their time on today will likely be automated by GenAI.

(They previously said 50%. This report was last updated in June of 2023, so we’d put that number at closer to 85% if we’re being honest.)

2 – GPTs are super simple 👍

Can you order DoorDash or talk to a friend on the phone? 

If so, you can build a GPT. 

The only real requirement (aside from basic communication skills) is a paid ChatGPT Plus account. 

From there, you can talk to the GPT Builder and just describe in plain English what you want it to build. 

Try this:

Yeah, we showed you the backend of building a GPT live today.

But, we also had a dedicated episode on building (and monetizing!) on GPTs hours after they were released. 

3 – Practice makes perfect

Does this whole, ‘spend 10 minutes on a GPT that will save you 10 hours’ spiel sound too good to be true? 

Well, here’s the part that is: simple GPTs sometimes produce simple results. 

Can you create a helpful GPT that saves you hours a week with only a few minutes of setup?

Absotootinglutely. 

Is it going to be perfect? No. 

Like Remy perfecting his ratatouille, practice makes perfect when it comes to optimizing GPTs. 

Try this: 

The great thing here? You can tell the GPT Builder in Builder mode how you want updated in your instructions based on the outputs in the Preview pane. 

Wanna take it a step further? 

Building GPTs requires some prompting basics. 

We cover that all the time in our free and freshly updated PPP course. 

Sound too good to be true? It’s not. 

This 75-minute course will take your prompting (and GPTs!) to the next level. 

⌚

Numbers to watch

$300 Million

Skild AI raises $300 million to build a general-purpose brain for Robots.

Now This …

Let us know your thoughts!

Vote to see live results

What task would you like to see us create a GPT for?

Winning option will become an AI in 5!

Login or Subscribe to participate in polls.

Reply

or to participate.