ChatGPT GPT 5 from OpenAI - AI for Complex Workflows UPDATED

Get up to speed on the August 2025 GPT-5 release. We’ll outline the unified system with two operating modes (Thinking and Flagship), explain how automatic switching works, and note what’s changed from earlier versions. You’ll also learn how legacy access works for models like GPT-4o, what may no longer appear by default, and how to enable older options when needed. By the end, you’ll understand the new lineup, the purpose of each mode, and how to select the right option for your workflow.

Tour the GPT‑5 interface and see where to select modes and models. We’ll clarify which plans (Free, Plus, Pro) include what modes, what “limited access” probably means for Free users, and how availability appears in the model selector. By the end, you’ll know exactly where to find the controls in the UI, what each plan provides, and how to choose the right mode for your work.

Learn how to tailor GPT-5 to your workflow. We’ll walk through Settings to adjust appearance and enable legacy models (including GPT-4o), then use Customize ChatGPT to define your assistant’s name, role, traits, and personality style (e.g., straight‑shooting, witty, or forward‑thinking). By the end, you’ll know how to save a personalized profile that makes the model respond more consistently with your preferences

This lecture clarifies how earlier OpenAI model families (o1, o3, and variants) map to GPT-5’s new Thinking and Flagship modes. Learn the logic behind model/mode mapping and avoid selection confusion. Focus on how these shifts impact practical usage and mode choice

A quick demonstration of the sheer intelligence of these models.

Understand the key differences between the o1-preview and the 01-mini model

The basics on setting up a ChatGPT account.  You can skip this lecture if you are already familiar with this and have started using ChatGPT.

In this video we look at how to use the ChatGPT o1 models for answering complex research questions and how this can save huge amounts of time over using the GPT 4o models.  The example chosen is a suitably complex question over the effect of tariffs on a country, and on global trade.  The lecture discusses the strengths and weaknesses of the different models.

Sample prompts, plus responses are given in the resources.  This is a shared ChatGPT page and isn't essential for the lecture - if you have difficulty accessing it, proceed to the next lecture.

This lecture leans in on the strengths of the o1-mini model to rapidly perform complex tasks.

Image creation is a matter of understanding the manner of prompting for images

The manner of editing images in Dall-E 3 isn't always obvious.  This video doesn't pretend that the process is perfect but instead shows methods of working around any issues.

With a strongly creative slant this exercise ends examining the very differing capabilites of the 4o and o1 models.  The lecture demonstrates the incredibly improve capability of the o1 models to respond to challenging prompts.  The shared link is for information only, if it doesn't work, proceed to the lecture.

This shows a creative method of working around some of the limitations of Large Language Models.

The tactics used here are great for encouraging creativity with these models and generalize outside of the ChatGPT environment.

Examining ways of creating preferences in how the models deal with responses, so that responses are reasonably tailored to your requirements.

Learn how the incredible reasoning power of o1 models can be used to create Youtube video timestamps from a transcript.

In this lecture we preview the immense coding capabilities of o1-mini.  This is not an indepth introduction, but we do cover the abilities of the model to work with HTML, CSS, JavaScript, and other more obscure methods.