Assumption is the mother of all screw-ups, Anonymous.

Introduction
AI generative chatbots are computer programs that use AI and machine learning to understand and respond to human input in a conversational, human-like way.
A prompt is a piece of input you give to an AI model —such as ChatGPT, Google Gemini, or DeepSeek — intended to trigger a specific type of output or accomplish a particular task. When carefully crafted, it clearly conveys your intent, allowing the AI to generate more accurate, relevant, and useful responses.
Prompt engineering has emerged as a critical skill in interacting with generative AI models, transforming vague queries into precise instructions that yield high-quality outputs.
Comprehensive Best Practices for Crafting Prompts
- Be clear and concise about you want. Ask Focused Questions: Keep your prompt brief and to the point. Vague prompts yield broad results; specificity narrows them. For example, instead of “What is AI?” ask about its applications in healthcare. Instead of “What is climate change?”, use “Summarize the impacts of climate change on agriculture in Sub-Saharan Africa”.
- Use clear language: Avoid jargon and vague terms. Use straightforward vocabulary, simple and clear language to ensure understanding.
- Define key terms: Clarify any ambiguous or context-dependent words.
- Provide context: Give the generative AI chatbot enough background information to guide its response. Tailor the context to your audience or purpose, such as specifying if it’s for students, hobbyists, or professionals. For educational purposes: “As a history teacher for 10th graders, explain the causes of World War II.”
- Specify the tone and style: Indicate the desired tone or style of the response (e.g., formal, informal or casual, friendly, humorous or playful, etc.) based on your needs. This helps in making the response approachable or professional as required. Example: “Write a formal email to request an extension on my project deadline,” “Respond in a playful, motivational tone like a fitness coach.”
- Use relevant keywords: Include relevant, domain terms or phrases to help the generative AI chatbot understand the topic. “Explain in detail the locus {z ∈ ℂ : |z − a| − |z − b| = c} for two fixed foci a, b ∈ ℂ and 0 < c < |a−b|. Cover both branches of the hyperbola and derive the implicit conjugate‐form in z and $\bar z$. Use precise mathematical language and include sketches of the transverse and conjugate axes.”
- Incorporate Examples: Include examples within your prompt to illustrate your points. This helps the AI understand your needs and structure its response effectively. Example for sentiment analysis: “Text: ‘I love this product!’ – Positive. Text: ‘It’s okay.’ – Neutral. Now analyze: ‘This is disappointing.’”
- Specify your desired response format, for example: bullet points, a step‑by‑step guide, or a conversational style. You can also request specific technical formats such as Markdown for structured text, JSON, LaTeX for mathematical notation, or code in HTML, Python, or other programming languages. This helps ensure the response is organized, styled, and delivered in the way that best fits your needs.
- Apply constraints to your request, for instance, you could limit the scope to a specific time period (such as “recent developments in the past two years”) or set a word, sentence, or character limits to ensure the response is focused and concise (“Respond in 150 words or less, emphasizing developments after 2023.”).
- Specificity: Provide the AI with precise details to ensure a clear understanding of your request, allowing for a relevant and accurate response. Include facts, figures, context, examples, or parameters that narrow the focus and minimize ambiguity.
- Use prompt templates for consistency and depth: include elements such as clarity (state your request plainly), context (provide background or purpose), examples (show what you expect), format, constraints (time frame, word limit), tone, feedback, and refinement (iterate for improvement). A well‑structured template keeps prompts comprehensive, focused, and reusable, e.g., [Context] + [Role] + [Task] + [Format] + [Constraints].
- Refine your prompts: Test and iterate (try different versions of your prompts to see which one yields the best results and adjust them based on previous responses); Provide feedback to refine your approach incrementally; Be Open-Minded (occasionally, the most valuable insights come from prompts that are slightly unconventional, creatively altered, or take an unexpected angle. Exploring these out‑of‑the‑box variations can spark responses you might not have anticipated, but which could prove surprisingly useful) and Use follow-up questions (ask additional questions to clarify or expand on the initial response).
CRISPE Framework is a structured method for crafting prompts:
- Context: Define the scenario (e.g., “You are a software engineer debugging a Python script”).
- Role: Specify the AI’s role, lens, or persona (e.g., “Act as a UX designer”).
- Instructions: Outline direct, unambiguous tasks (e.g., “List three solutions to fix the error: ‘Index out of range’”).
- Style: Define the tone and format of the output (e.g., “Write in a step-by-step tutorial style”).
- Examples: Provide a model response (both ideal and wrong examples) if applicable to fine-tune output.
By carefully balancing these elements, you create prompts that encourage precise, relevant, and high-quality responses from a generative AI chatbot. Ongoing experimentation with various structures, templates, and stylistic variations will help you refine your approach, uncovering the formats that work best and are most effective for achieving your specific goals.
AI Chatbots
AI chatbots offer assistance with queries, coding, and problem-solving. Here are some leading options:
- Kimi: A conversational AI for general knowledge and creative tasks 🌟.
- Qwen 3: A versatile AI for multilingual conversations and complex problem-solving 🌟.
- DeepSeek: Optimized for technical queries, particularly coding and data analysis 🌟.
- Gemini: Google’s AI chatbot for research and productivity tasks Google integrations (Docs, Search). 🌟
- ChatGPT: OpenAI’s widely-used conversational model for diverse applications 💰.
- Perplexity: An AI-powered search tool providing concise, sourced answers.
- Grok: xAI’s chatbot for insightful answers and real-time information, accessible via grok.com and other platforms.
- LMArena: An open platform for evaluating AI through human preference. Select Direct Chat.
- Claude: Anthropic’s conversational AI, known for safety and ethical responses.
- ERNIE is China’s answer to ChatGPT. It is an advanced, open-source, AI-based chatbot that can generate text, images, and videos.
- Poe: A platform to interact with multiple AI models, including custom bots.
- Yupp gives you two or more responses from some of the best AIs, side by side.
- Z.ai, free AI Chatbot powered by GLM-4.5.
- MetaAI is a free conversational assistant embedded across Facebook, Instagram, WhatsApp and Messenger. It answers natural-language queries, offers follow-up dialogue, identifies and edits image content, and generates visuals on demand.
Search Engines and AI-Powered Search
Search engines with privacy features or AI enhancements can improve your online experience:
- Google AI Mode: Integrates AI for enhanced search results and summaries.
- Bing: Microsoft’s search engine with integrated Copilot AI for interactive queries.
- Copilot: An AI assistant integrated into Bing for task automation and insights.
- SearXNG: An open-source metasearch engine that aggregates results while prioritizing privacy.
- Phind: An AI-driven search engine tailored for developers and technical queries.
- DuckDuckGo: A privacy-focused search engine that blocks trackers and doesn’t store personal data. Its “bang” feature allows direct searches on specific sites using commands like
!g (Google), !w (Wikipedia), or !yt (YouTube). For example, searching !yt tutorial takes you directly to YouTube search results for “tutorial.”
- Startpage: Delivers Google’s search results anonymously, stripping personal data for privacy.
- Mojeek: A privacy-focused search engine with its own index, independent of Google or Bing.
- Qwant: A European privacy-first search engine emphasizing user data protection.
- Opera (Aria): Opera’s built-in AI assistant for browsing and task assistance.
- Brave (Brave Leo AI): Brave’s AI chatbot integrated into its browser for privacy-focused assistance.