The Complete Obsolete Guide to Generative AI

The Complete Obsolete Guide to Generative AI
Copyright
contents
front matter
   foreword
   preface
   acknowledgments
   about this book
      About the code
      liveBook discussion forum
   about the author
   about the cover illustration
1 Understanding generative AI basics
   Stepping into the generative AI world
   Categorizing AI models by function and objective
      Understanding usage tokens
      GPT-4 models
      GPT-3.5 models
      GPT-3 models
   Model fine-tuning
   The technologies that make generative AI work
   AI and data privacy and ownership
   AI and reliability
   What’s still ahead?
   Summary
2 Managing generative AI
   Accessing GPT models
   Learning by playing
   Accessing Python code samples
   Accessing curl code samples
   Completion configurations
      Setting the mode
      Setting the temperature
      Setting the Top P value
      Working with stop sequences
      Setting the frequency penalty
      Setting the presence penalty
      Working with Best Of
      Working with the Inject Start Text setting
   Summary
3 Creating text and code
   Automating accuracy checking
   Creating new contextually aware content
      Setting up your environment for Python
      Creating your prompt (using Python)
   Generating specialized documents
   Generating programming code
   Interactive coding with Copilot
   Try this for yourself
   Summary
4 Creating with media resources
   Generating images
      Providing detailed prompts
      Prompting for images
   Generating video
      AI-assisted video editing
      Text-to-video slide shows
   Generating presentation resources
   Generating voice
   Audio transcriptions
   Generating music
   Try this for yourself
   Summary
5 Feeding data to your generative AI models
   Indexing local data archives
   Seeding a chat session with private data (ChatPDF)
   Connecting your AI to the internet (Auto-GPT)
   Try this for yourself
   Summary
6 Prompt engineering: Optimizing your experience
   What is prompt engineering?
   Prompt engineering best practices
      Be specific
      Be clear
      Avoid unnecessary words
      Separate reference text from your instructions
      Be positive, not negative
      Control for temperature (randomness)
   Zero-shot and few-shot prompting
   Prompt for time-series data: A practical example
      Visualizing the data
      Graphing the time-series data without normalization
      Graphing the time-series data with normalization
   Try this for yourself
   Summary
7 Outperforming legacy research and learning tools
   Asking for investment guidance
   Connecting search engines to AI using LangChain
   Using LangChain to analyze multiple documents
   Teaching yourself to program, to speak a new language, or anything else
   Integrating LLMs into your daily work
      Spreadsheet integration
      Kanban integration
      Slack integration
      Salesforce integration
      Code version control
      Photoshop integration
   Try this for yourself
   Summary
8 Understanding stuff better
   Using GPT to replace analytics
   Using GPT to replace sentiment analysis
      Some background to sentiment analysis
      Testing sentiment analysis through GPT
   Try this for yourself
   Summary
9 Building and running your own large language model
   Some background to building your own model
   Selecting a base LLM model for configuration
   Configuring and building your model
   Fine-tuning your model
      Creating a dataset
      Training your model
   Creating your own GPT
   Summary
10 How I learned to stop worrying and love the chaos
   What the workers of the world can reasonably expect
   What your next business startup will look like
   Artificial general intelligence: Where it’s all going
   Should AI be regulated?
   The road ahead
      Quantum computing
      Neuromorphic computing
      Advanced hardware acceleration
      Reinforcement learning and meta-learning
      Multimodal learning
      Explainability and interpretability
      Data efficiency and few-shot learning
      Domain-specific knowledge integration
   Second-order effects
      Investment markets
      Human innovation
      Employment markets
      On-demand media
      On-demand journalism
   Summary
11 Experts weigh in on putting AI to work
   Including projects discussed in your book, where have you had the greatest success applying AI to solving practical problems?
      Daniel Sanz Becerril
      Leo Porter and Daniel Zingaro
      Chrissy LeMaire
      Paul McFedries
   What would you say is the most transformational generative AI use case right now in your corner of the IT world?
      Chrissy LeMaire
      Daniel Sanz Becerril
      Leo Porter and Daniel Zingaro
      Paul McFedries
   What’s the next big thing you see AI bringing to the world that very few people yet anticipate?
      Leo Porter and Daniel Zingaro
      Daniel Sanz Becerril
      Paul McFedries
      Chrissy LeMaire
   What’s the most fun you’ve ever had interacting with generative AI?
      Chrissy LeMaire
      Paul McFedries
      Daniel Sanz Becerril
      Leo Porter and Daniel Zingaro
   What was your most spectacular generative AI-fueled disaster and/or disappointment?
      Chrissy LeMaire
      Leo Porter and Daniel Zingaro
      Nathan Crocker
      Daniel Sanz Becerril
      Paul McFedries
   Do you see recent reports about GitHub Copilot losing significant money on each monthly account as indications that we haven’t yet figured out a sustainable gen AI model, or is it just a temporary blip?
      Daniel Sanz Becerril
      Chrissy LeMaire
      Paul McFedries
      Leo Porter and Daniel Zingaro
   Could you share some of the AI tools you’re using. Do you have any tips or warnings for readers experimenting with generative AI?
      Paul McFedries
      Chrissy LeMaire
      Leo Porter and Daniel Zingaro
      Nathan Crocker
Appendix A. Important definitions and a brief history
   Some critical AI definitions
   A very (very) brief history of AI
Appendix B. Generative AI resources
   General LLM interaction tools
   AI application development platforms
   Third-party tools
   Writing tools
   Image generation
   Data analytics
   Investment and financial
   Speech-to-text
   Text-to-speech
   Text-to-music
   Text-to-video
   Text-to-video presentations (including animated and lifelike avatars)
   Slide deck generation
   Text, audio, and video language translation
   Domain specific
Appendix C. Installing Python
   Installing Python on Windows
   Installing Python on macOS
   Installing the pip Python package manager on Linux
index