Writing With Generative AI

Generative AI can be a powerful tool for creating content. However, a more responsible use is as a writing assistant. Generative AI can be highly valuable to aid a writer in tasks such as proofreading, editing and translating languages.

Proofreading With Generative AI - Syntax vs Semantics

Traditional spelling and grammar checking software can only find syntax errors, where Generative AI can find syntax errors in addition to semantic errors.

If you are not familiar with syntax and semantics, consider this example:

Syntax & Semantics
  1. "The house is in the cat." - This sentence has correct syntax, the spelling and grammar follow English rules. However, it doesn't make sense. Syntax relates to rules.
  2. "The cat is in the house." - This sentence has correct syntax and semantics. It is grammatically correct and makes sense. Semantics relates to meaning.
Writing With Generative AI - Responsibly

Generative AI should be used to empower writers, not replace them. It is important to use AI responsibly. We asked the Generative AI, Claude to summarize some principles to follow, and agree with this response:

Here are some principles for the responsible use of AI writing assistants:

Attribution
If an AI tool generates content that gets published or shared in any form, always disclose that the writing is AI-generated. Not doing so could constitute plagiarism or misrepresentation.
Limitations
Understand the limitations around AI comprehension and originality compared to human understanding. Fact check content rigorously and do not blindly rely on AI writing.
Augmentation
Use AI tools to augment rather than replace your own critical thinking and writing skills through proofreading, expanding ideas, overcoming writer’s block etc. But develop your own talent.
Bias Mitigation
Choose writing assistants carefully to minimize harmful societal biases around race, gender, culture etc reflected in output text. Audit content thoroughly.
Legality
Ensure text complies fully with copyright laws and does not closely mimic or steal existing written works without attribution, which could constitute plagiarism.
Ethics
Do not request or include inappropriate, dangerous or illegal content. Report concerns over AI biases or capabilities responsibly.
Accessibility
Advocate for and enable writing solutions fairly across student populations as AI penetrates academia, maintaining equitable standards.

The core idea is using AI writing tools thoughtfully and transparently as assistants rather than as unsupervised black boxes. The human should remain accountable for final output quality, biases and ethics.

Generative AIs We Use

Here is a list of generative AI's we frequently use, along with their provided descriptions in quotes, and our comments:

books
  • Gemini - "Try Gemini, Google's helpful collaboration chat tool, and use generative AI to boost productivity, enhance creativity and bring your ideas to life."
    Great for grammar questions, quick topic research, and small proofreading tasks
  • Claude - "Claude.ai is currently in open beta and usage of the platform may be limited for unpaid users. Claude for Business. Claude is a next-generation AI assistant for your tasks, no matter the scale. Our API is currently being offered to a limited set of customers and researchers."
    A good starting point for summarizing topic research with responses in list form
  • Chat GPT - "ChatGPT is a free-to-use AI system. Use it for engaging conversations, gain insights, automate tasks, and witness the future of AI, all in one place. ChatGPT is an AI-powered language model developed by OpenAI, capable of generating human-like text based on context and past conversations.
    Provides an API useful for proofreading entire manuscripts
  • Preplexity AI - "Perplexity AI unlocks the power of knowledge with information discovery and sharing."
    Helpful for academic writing and citing sources as the AI provides references with responses

Image from CLEAN PNG

Prompt Engineering

The ability to write great prompts provides a huge advantage when using generative AI. Prompt engineering is the process of structuring text to improve generative AI responses towards a goal. Here are some resources for learning how to write quality prompts: