How to use ChatGPT and Generative AI to help create content

Master auto-regressive image generation with ChatGPT-4o to create consistent, editable images and infographics with inline image editing by iterating left-to-right, top-to-down, and building on prior context, unlike diffusion models.

Learn to generate images with Gemini in the Google app using prompts like 'can you make me a picture of' and use keywords such as dslr, high-contrast, and 4k.

Explore Google Vertex AI Studio's media suite for generating images, audio, music, and video. See Veo video generation, Chirp text-to-speech, Lyra sound, and Imagine 3 image tools.

Explore auto regressive image generation models and the GPT-4o image generation approach, compare left-to-right creation with diffusion, and learn how context and text accuracy improve consistency and cost.

Explore generating professional infographics with Nano Banana Pro, a Google image model, using reference images, nine-panel layouts comparing LEO, MEO, and GEO, with guidance options and explain like a 13-year-old.

Generate text effects with Adobe Firefly by entering text and describing the desired look, using prompts to create letters or words with fire, water, or peacock feather styles.

Explore Adobe Firefly's generative recolor to recolor vector logos and SVG artwork, experiment with color palettes and prompts, and apply seasonal themes like holidays and summer.

Case study reveals Coca-Cola's use of Midjourney to create an AI-generated ad by editing each video frame and recombining them into a cohesive film.

Demystify large language models by explaining how LLMs infer the next word from probabilities, the transformer background, and the role of parameters and cost, while noting their tendency to hallucinate.

Explore the basics of prompt engineering to optimize prompts for large language models like chatgpt. Learn techniques that improve text summarization, information extraction, question answering, and coding tasks.

Explore zero shot and few shot prompting in ChatGPT and generative AI, showing how zero shot uses prebuilt knowledge and few shot uses examples, with limitations in reasoning and numbers.

Learn to evaluate models and prompts with evals scoring outputs against a rubric. Identify input, rubric, and response, and compare prompts using multiple choice, exact match, or rubric-based evaluations.

Learn a practical framework for writing powerful prompts, including context, tone, background data, task rules, and examples, and how to test and refine prompts for production while handling edge cases.

Prime prompts by framing context and adopting expert roles to elicit specialized, higher-quality AI responses. See how context shifts produce tailored outputs for cover letters, business plans, and executive guidance.

Use chain of thought prompting to reveal the model’s work for better logical and math reasoning. Harness few-shot and zero-shot prompts with context and step-by-step thinking for more accurate answers.

Learn how grounding large language models improves accuracy by supplying task-specific data via retrieval and augmented generation, and compare grounding with fine-tuning for context using embedding and a vector database.

Explore multimodal LLMs that process images, audio, and video via alignment and modality modules, converting media into tokens the LLM understands; examine advantages and use cases.

Discover how retrieval-augmented generation boosts LLM accuracy by fusing internal data with external sources and citing sources. Implement it with five lines of code and private data use cases.

Discover the model context protocol (MCP) as an open standard that wires AI applications with tools, resources, prompts, and hosts via a USB-C hub, enabling interoperable, vendor-neutral collaboration.

Generalize your llm knowledge to new chatbots and future releases by mastering prompting principles, one-shot and multi-shot prompting, chain-of-thought techniques, and clear instructions, with awareness of hallucinations.

Highlighting ChatGPT hallucinations that produced fictitious legal citations in a federal New York filing, the case led to fines and underscored the need to ground and verify outputs.

Learn to generate code with ChatGPT using prompts like 'write me code' in a chosen language. See how to parse addresses and phone numbers across Python, JavaScript, C++, and Perl.

Learn to use chatgpt as an ideas assistant for crafting watch-focused blog articles, from priming prompts to expanding topics and refining content within a $2,000 budget.

Integrate the OpenAI API into an app using text completion, create an API key, install the OpenAI Python library with pip, and perform a completions.create call with model and parameters.

Use ChatGPT's code interpreter in beta to upload CSV data, write Python code with pandas, and derive census insights like population from the dataset.

Compare major LLM options such as Bing Chat, Claude, ChatGPT, Bard, and Google Search Experience to identify ideal use cases for up-to-date info, copywriting, and PDF interactions.

Define AI agents as software that interact with the environment, collect data, and steer self-determined tasks toward goals using multiple models and a vector memory store.

This case study shows how Buzzfeed uses ChatGPT to generate on-the-fly content from quiz responses. It builds prompts from user answers and queries the ChatGPT API for personalized ideas.

Build a custom GPT with the GPT Builder, uploading your data and instructions to create a shareable chatbot like Calorie Companion.

Explore the 2024 Gemini update in Bard, featuring a new ai model, multimodal image analysis, extensions for live data, and built-in fact-checking for content creation.

Chat with a PDF using Claude to extract lease details—start and end dates, rent, late fees, and no-pets terms—while exploring lease-breaking and section O.

Use ChatGPT to generate compelling website marketing copy and headlines, refine iterations, and speed up building sites for selling products, with examples like phone camera pitches and Samsung-style copy.

Discover the do's and don'ts of generative AI, including benefits like original content, art inspiration, quick prototyping, and awareness of copyright and plagiarism risks.

Learn how fine tuning customizes pre-trained models like GPT-3.5 for specific use cases, including customer support and niche translations, using OpenAI APIs and a labeled fine-tuning set.

Set up your python environment using either a local install on Mac, Windows, or Linux, or a browser-based Replit workspace, then install the OpenAI library with pip.

Learn OpenAI API pricing and usage limits, compare GPT-4 and GPT-3.5 Turbo, and set approved, hard, and soft usage limits using token-based costs on the dashboard.

Update on the OpenAI API notes the Completions API deprecation and shift to chat completions, with GPT-3.5 Turbo Instruct as the supported completion model and DaVinci deprecated.

Learn how to use the completions API for translating articles with zero-shot and few-shot prompting, including prompt design, token considerations, and cost-aware performance.

Upgrade your OpenAI API project to the new version with the OpenAI migrate tool, which automatically updates Python files and completion calls to the new API signature.

Learn to build a chatbot with the OpenAI chat completion API by using system and user prompts, context from prior messages, and managing context windows.

Generate phone wallpapers using the OpenAI image generation API with Dall-E three and Dall-E two, testing resolutions, HD and standard qualities, and in-house prompts.

Watch a chat completion demo in the OpenAI API, showing how environment variables, base messages, and few-shot encoding establish an initial prompt and shape a chatbot’s trajectory through a loop.

Demonstrate how to transcribe blog post text into speech audio using the OpenAI text-to-speech API, selecting models and voices in Python to generate podcast-ready mp3 files.