Название: Visualizing Generative AI: How AI Paints, Writes, and Assists (Early Release)
Автор: Priyanka Vergadia, Valliappa Lakshmanan
Издательство: O’Reilly Media, Inc.
Год: 2025-06-04
Страниц: 120
Язык: английский
Формат: epub
Размер: 10.1 MB

Generative AI has the potential to innovate and evolve business processes, but workers are still figuring out how to build with, optimize, and prompt GenAI tools to fit their needs. And of course, there are pitfalls to avoid, like security risks and hallucinations. Getting it right requires an intuitive understanding of the technology's capabilities and limitations. This approachable guidebook helps learners of all levels navigate GenAI—and have fun while doing it.

Generative AI, a subset of Artificial Intelligence, has taken the world by storm. Unlike traditional AI, which analyzes and interprets data, Generative AI creates something completely new, such as text, images, or music. Its rapid adoption, impressive capabilities, and potential for revolutionizing industries have made it a hot topic of discussion.

One of the key reasons for Generative AI’s popularity is its user-friendliness. Anyone can generate text, images, or music with just a few clicks, making it accessible to a wide range of people, both tech-savvy and non-tech-savvy alike. Another reason for its popularity is its speed. Generative AI can generate high-quality content in a matter of seconds, which saves users a significant amount of time. Finally, Generative AI is constantly improving. With each passing day, it becomes more sophisticated, capable of producing more realistic and creative content.

The potential for Generative AI is enormous. It has the potential to revolutionize industries such as customer support, marketing, and entertainment. For example, Generative AI can be used to create personalized customer support experiences, generate high-quality marketing content, and create realistic visual effects for movies and TV shows. However, as with any technology, Generative AI also brings unique risks. For instance, it can be used to create fake news, deep fakes, and other forms of misinformation, which could lead to serious consequences. Additionally, the widespread adoption of Generative AI could lead to job displacement, as more and more tasks become automated.

Generative AI (GenAI) is a form of machine learning, because the AI is taught using large amounts of data, but its output tends to be much more open-ended than that of most ML applications. The output of GenAI models can be videos, audios, etc. Large language models (LLMs) are GenAI models that are trained on textual inputs to generate natural language text. ChatGPT is an example of a GenAI product that is powered by an LLM.

At the time of writing, the API keys can be created at the following links:
• OpenAI Platform
• Gemini: API Key in Google AI Studio
• Groq (for open models such as Llama3 and Mixtral): API Keys
• Anthropic Console

Now, make sure that you can load these keys using the code stored in the repository, in the file invoke_llm.py. While each of the four providers above has a Python API, the APIs are all slightly different. To make it convenient to try out all four (and avoid getting locked into any one), we’ll use the Python package LangChain in this section.