AI Tools and Applications UNIT-I

 

UNIT – I

UNDERSTANDING AI AND AI FOR DOCUMENTS AND PRESENTATIONS

Core Content

What is AI: AI vs. ML vs. Deep Learning vs. Generative AI; brief history (Turing Test → ChatGPT → generative AI era); how LLMs work (training data, next-token prediction, temperature, hallucinations, context windows, knowledge cutoffs). AI ecosystem: ChatGPT, Claude, Gemini, Microsoft Copilot strengths and differences. AI for presentations: Gamma.app — full slide deck from a paragraph prompt; Microsoft Copilot in PowerPoint — slide generation, design suggestions, speaker notes; Canva AI Magic Design, text-to-presentation, brand kits; Beautiful.ai smart slide layouts; comparing AI-generated vs. manually created presentations. AI for document writing: Microsoft Copilot in Word — draft, rewrite, summarize; Gemini in Google Docs — writing assistance and inline Q&A; Notion AI — meeting notes, project plans, knowledge base; Claude Projects — organizing documents and having a persistent AI assistant over a set of files.

AI Integration

Use Gamma.app to generate a presentation on an engineering topic in under 2 minutes; use Copilot in Word or Gemini in Docs to draft and improve a report; compare ChatGPT, Claude, Gemini, and Microsoft Copilot on the same writing task; use Grammarly to review your own writing.

AI tools

AI tools : are software applications powered by artificial intelligence that perform specific tasks like writing text, generating images, analyzing data, or writing code.

What is an AI Tool?

An AI tool uses machine learning, natural language processing (NLP), or computer vision to process inputs and produce automated outputs.

·         Content Generation: Tools like ChatGPT or Claude write essays, emails, and code.

·         Information Retrieval: Tools like Perplexity fetch real-time facts and cite web sources.

·         Media Creation: Systems like text-to-image generators create visual designs or clone voices

AI applications

AI applications: refer to the broader implementation of AI technology to solve real-world problems across industries.

What is an AI Application?

An AI application represents the functional use of AI within a domain or workflow to automate and improve efficiency.

·         Customer Service: Virtual agents and chatbots handle routine consumer requests 24/7.

·         Healthcare: AI models assist with diagnostic imaging analysis and patient data tracking.

·         Recommendation Systems: Streaming platforms and e-commerce sites suggest products or media based on past user behavior

Artificial Intelligence (AI)

Artificial Intelligence (AI) is the broad field of making computers or machines perform tasks that normally require human intelligence.

How AI Works

• Learning from data: AI systems look at large amounts of data to find patterns instead of following a fixed list of rules.
• Improving over time: The system gets better at making predictions or decisions the more data it processes.
• Using math: AI uses advanced math and fast calculations to guess the right output, like the next word in a sentence or an object in a photo.

Common Types of AI

• Narrow AI: Systems built to do one specific job, like spam filters, chess programs, or facial recognition.
• Generative AI: Tools that create new content like text, images, music, or code based on user prompts.
• Machine Learning: A core part of AI where computers learn from past data without being explicitly programmed for every choice.

Where You See AI Every Day

• Navigation: Apps like Google Maps use AI to find the fastest routes.
• Virtual assistants: Tools like Siri, Alexa, and Gemini help answer questions and manage tasks.
• Recommendations: Streaming services and online stores use AI to suggest movies or products you might like.

These tasks include:

• Learning
• Reasoning
• Problem-solving
• Understanding language
• Recognizing images
• Making decisions

Example:
A Google Maps app suggesting the best route is using AI to help make a decision.

Simple definition:

AI is the ability of a computer or machine to perform tasks that normally require human intelligence.

 

Machine Learning (ML)

Machine Learning is a part of AI.

Instead of giving the computer every rule, we provide it with data and examples, and the computer learns patterns from that data.

Machine learning is a subfield of artificial intelligence that uses algorithms and data to train computers to learn, predict, and make decisions without being explicitly programmed.

How Machine Learning Works

• Data Input: You feed raw or processed data into an algorithm.
• Training Process: The computer looks for patterns, trends, and rules in the data.
• Model Creation: The algorithm becomes a trained model that can evaluate new, unseen information.

 

Example: Spam Email Detection

Suppose we give a computer:

• 1,000 spam emails
• 1,000 normal emails

The ML system studies their patterns.

Later, when a new email arrives, it predicts:

"Spam" or "Not Spam"

 

Simple definition:

Machine Learning is a method of AI in which computers learn patterns from data and use them to make predictions or decisions.

 

Deep Learning (DL)

Deep Learning is a part of Machine Learning.

It uses special models called neural networks, which are inspired by the way the human brain processes information.

 

Deep learning is a specialized branch of machine learning that uses multilayered artificial neural networks to mimic the human brain when processing data and making decisions

Deep Learning is especially useful for:

• Images
• Speech
• Video
• Natural language
• Complex pattern recognition

Example: Face Recognition

When you unlock your phone using your face:

Camera → Image → Deep Learning model → Face recognized → Phone unlocked

The model learns from many examples of faces.

 

Simple definition:

Deep Learning is a type of Machine Learning that uses multi-layered neural networks to learn complex patterns from large amounts of data.

 

Generative AI

Generative AI is a type of artificial intelligence that creates new content, such as text, images, music, and code, by learning patterns from existing data.

 

How It Works

• Training: Models study vast amounts of data using deep learning and neural networks to understand underlying structures and rules.
• Prompts: Users give instructions or questions in normal, everyday language.
• Prediction: The system predicts and generates a brand-new, unique response based on those learned patterns.
• Foundation Models: Large systems like ChatGPT or Gemini serve as the base brain for these tools.

 

Common Uses

·         Text: Writing emails, summarizing long documents, and generating software code.

·         Images & Video: Creating digital artwork, realistic photos, and special effects.

·         Audio: Composing original music and synthesizing natural-sounding speech.

 

Generative AI is a type of AI that can create new content.

Instead of only classifying or predicting something, it can generate something new.

It can create:

• ✍️ Text
• 🖼️ Images
• 🎵 Music
• 🎬 Videos
• 💻 Computer code
• 🎙️ Speech

Examples

ChatGPT → Generates text and answers questions.

Image-generation AI → Generates images from a text description.

AI coding assistant’s → Generate computer programs.

Simple definition:

Generative AI is AI that learns patterns from existing data and generates new content such as text, images, audio, video, or code.

AI vs ML vs Deep Learning vs Generative AI

The easiest way to remember the relationship is:

AI → Machine Learning → Deep Learning

And Generative AI can use deep-learning models to create new content.

 

Note: Generative AI is not simply "the next level" of Deep Learning in every case. It is a category of AI systems focused on generating content, and many modern generative systems are built using deep learning.

 

 

Technology	What does it do?	Simple example
AI	Makes machines intelligent	Smart assistant
ML	Learns from data	Spam detection
Deep Learning	Learns complex patterns using neural networks	Face recognition
Generative AI	Creates new content	ChatGPT generating an answer

 

Brief history (Turing Test → ChatGPT → generative AI era);

Can machines think? → Can machines learn? → Can machines understand language? → Can machines create?

Alan Turing and the Turing Test – 1950

The Turing test is a method created by mathematician Alan Turing in 1950 to see if a machine can show human-like intelligence through conversation

In 1950, British mathematician Alan Turing asked an important question:

“Can machines think?”

He proposed the Turing Test.

What is the Turing Test?

A human communicates with:

• another human
• a computer

Without knowing which one is the computer.

If the human cannot reliably tell whether the responses come from a machine or a human, the machine is considered to have demonstrated human-like conversational ability.

Example:

Human → Question → Computer

Human ← Answer   ← Computer

 

Can the human identify that it is talking to a machine?

How It Works

• Three players: A human judge, a second human, and a machine.
• Text chat: The judge talks to both players using a keyboard and screen without seeing them.
• The goal: The judge asks questions to figure out which player is the machine and which is the human.
• Passing the test: The machine passes if the judge cannot tell it apart from the human.

Key point: Turing's work became an important foundation for thinking about machine intelligence.

 

Birth of Artificial Intelligence – 1950s

The term Artificial Intelligence became established as a field during the 1950s.

Early AI systems mainly used:

• Rules
• Logic
• Search
• Problem-solving techniques

Example

A computer could be programmed with rules such as:

IF temperature > 38°C

THEN fever = YES

This is called rule-based AI.

The computer follows instructions given by humans.

Machine Learning – 1980s–2000s

AI gradually moved from:

"Tell the computer every rule"

To:

"Give the computer data and let it learn patterns."

This led to the growth of Machine Learning (ML).

Example

Instead of programming:

How LLMs work (training data, next-token prediction, temperature, hallucinations, context windows, knowledge cutoffs).

LLM = Large Language Model

An LLM is an AI system that can understand and generate human-like text. Examples include ChatGPT, Gemini, Claude, and others.

With advancements in artificial intelligence, NLP and LLMs have become essential for enabling machines to understand and interact with human language.

• Natural Language Processing (NLP): Focuses on analyzing and interpreting text or speech, powering applications like sentiment analysis, machine translation and speech recognition

• Large Language Models (LLMs): Extend NLP capabilities by learning from large datasets to generate human-like text, answer questions and perform reasoning tasks

A simple way to remember how an LLM works:

Training Data → Learning Patterns → Next-Token Prediction → Generated Response

 

Training Data

             

                           

 

An LLM learns from a very large collection of text and other data.

Training data may include:

• Books
• Websites
• Articles
• Documents
• Code
• Other publicly available or licensed data

The model does not simply memorize everything like a database. During training, it learns patterns and relationships between words, phrases, and concepts.

Example

Suppose the training data contains:

"The sun rises in the ___."

The model learns that "east" is strongly associated with this sentence.

Simple idea

Training = Learning patterns from a huge amount of data.

 

Next-Token Prediction

This is one of the most important ideas behind LLMs.

           

An LLM generates text one token at a time.

A token can be:

• A complete word
• Part of a word
• Punctuation
• A number or symbol

Example

Input:

The sky is

The model predicts possible next tokens:

Possible token	Probability
blue	High
cloudy	Medium
beautiful	Lower
green	Very low

 

The model might choose:

The sky is blue

Then it predicts the next token again.

So:

The → sky → is → blue →..........

Important

LLMs don't usually "write the whole sentence at once."

They repeatedly perform:

Look at context → Predict next token → Add token → Predict again

 

Temperature:

Temperature controls how predictable or creative the model's output is.

Think of temperature like a creativity control.

Low temperature

The model tends to choose highly probable tokens.

Result: More predictable and focused answers.

Example:

"The capital of India is..."

Likely answer:

New Delhi

Higher temperature

The model is more willing to select less-probable possibilities.

Result: More varied or creative responses.

Temperature	Output
Low	More predictable
Medium	Balanced
High	More creative/random

 

Remember: Temperature does not give the model new knowledge. It changes how it selects among possible outputs.

 

 

Hallucinations

An AI hallucination happens when an LLM produces information that sounds correct but is actually incorrect, made up, or unsupported.

 

Example

Student asks:

"Who wrote a book that does not exist?"

The AI might sometimes give:

"The book was written by Dr. XYZ in 1998."

It may sound convincing, but the person or book could be completely fictional.

Why does this happen?

Because an LLM is fundamentally trained to predict likely sequences of tokens. It is not automatically a perfect fact-checking system.

Context Window

The context window is the amount of information an LLM can consider at one time while generating a response.

Think of it as the model's working space for the current conversation or input.

Example

You give the model:

"My name is Ravi. I am studying CSE. I like Python."

Then ask:

"What programming language do I like?"

The model can use the earlier information in its available context and answer:

Python

Simple analogy

Imagine a student writing an answer while looking at a limited number of pages on a desk.

Context window = information currently available on the desk.

Knowledge Cutoff

A knowledge cutoff is the point in time after which information may not have been included in a model's original training.

Example

Suppose an AI model's training knowledge ends around 2025.

If you ask:

"What happened yesterday?"

The model may not know the answer from its original training.

This is because:

Training data → Knowledge available to model

Information created after the training cutoff may not be known unless the AI has access to tools such as web search or other updated data sources.

AI ecosystem: ChatGPT, Claude, Gemini, Microsoft Copilot strengths and differences

The modern AI ecosystem includes several powerful generative AI platforms. Among the most widely used are ChatGPT, Claude, Gemini, and Microsoft Copilot. Although all can generate text, answer questions, summarize information, write code, and analyze content, they differ in their strengths, integrations, workflows, and ideal use cases.

ChatGPT

Developed by: OpenAI

Key Strengths

• Strong general-purpose AI assistant
• Excellent for reasoning, explanation, brainstorming, and writing
• Strong coding and programming assistance
• Useful for data analysis and visualization
• Can work with documents, images, and other types of content
• Supports customized workflows and AI-powered tools
• Useful for research, education, content creation, and professional work
•  

Common Applications

• Academic research and literature analysis
• Lesson planning and teaching materials
• Programming and debugging
• Data analysis
• Report and presentation preparation
• Brainstorming and creative writing
• Business analysis

Example

A teacher can use ChatGPT to convert a syllabus into a lesson plan, create quiz questions, explain difficult concepts, and prepare presentation content.

Claude

Developed by: Anthropic

Key Strengths

• Strong long-document analysis
• Good at handling large amounts of textual information
• Strong writing and editing
• Useful for analyzing policies, reports, papers, and contracts
• Strong emphasis on careful and structured responses
• Particularly useful for document-heavy professional workflows
•  

Common Applications

• Research-paper analysis
• Summarizing lengthy documents
• Editing academic and professional writing
• Requirements analysis
• Business documentation
• Code review

Example

A researcher can provide a long research document and ask Claude to identify themes, summarize arguments, and organize findings.

Google Gemini

Developed by: Google

Key Strengths

• Strong connection with the Google ecosystem
• Useful for multimodal tasks involving text, images, and other content
• Integration with Google productivity services can be valuable
• Useful for research and information discovery
• Strong capabilities for coding and technical tasks
• Particularly useful for users already working extensively with Google services

 

Common Applications

• Google Workspace productivity
• Research and information gathering
• Document and email assistance
• Coding
• Image and multimodal analysis
• Education and content creation

Example

A student or teacher using Google Workspace can use Gemini alongside documents, emails, presentations, and other Google productivity workflows.

Microsoft Copilot

Developed by: Microsoft

Key Strengths

• Deep integration with the Microsoft ecosystem
• Useful with Microsoft 365 applications
• Supports workplace productivity
• Can assist with documents, spreadsheets, presentations, emails, and meetings
• Useful for enterprise environments
• Strong connection to Microsoft's business and cloud ecosystem

Common Applications

• Microsoft Word document assistance
• PowerPoint presentation creation
• Excel-related analysis
• Outlook email assistance
• Teams meeting and collaboration workflows
• Enterprise knowledge work

Example

An employee can use Copilot within Microsoft 365 to help draft a Word report, summarize a Teams meeting, prepare an email, or work with information in Excel.

 

 

 

 

 

Feature	ChatGPT	Claude	Gemini	Microsoft Copilot
General-purpose AI	⭐ Strong	⭐ Strong	⭐ Strong	⭐ Strong
Writing	Strong	Very strong	Strong	Strong
Long documents	Strong	Very strong	Strong	Strong
Coding	Very strong	Very strong	Very strong	Strong
Data analysis	Very strong	Strong	Strong	Strong
Multimodal tasks	Strong	Strong	Very strong	Strong
Google integration	Limited compared with Gemini	Limited	Excellent	Limited
Microsoft 365 integration	Limited compared with Copilot	Limited	Limited	Excellent
Academic work	Very strong	Very strong	Strong	Strong
Enterprise productivity	Strong	Strong	Strong	Very strong
Creative tasks	Very strong	Strong	Strong	Strong

ChatGPT

"Help me think, create, analyze, code, and solve problems."

Claude

"Help me understand and work with large amounts of information and text."

Gemini

"Help me work with AI and Google's ecosystem."

Microsoft Copilot

"Help me get work done inside Microsoft's productivity ecosystem."

AI for presentations: Gamma.app — full slide deck from a paragraph prompt; Microsoft Copilot in PowerPoint — slide generation, design suggestions, speaker notes;

What is AI for Presentations?

AI for presentations means using Artificial Intelligence tools to help create presentation slides quickly and effectively.

Instead of creating every slide manually, we can give an AI tool a topic or paragraph prompt, and it can generate:

• Slide titles
• Content and bullet points
• Images and visual elements
• Slide layouts
• Speaker notes
• Presentation structure

Two useful tools are Gamma.app and Microsoft Copilot in PowerPoint.

Gamma.app

Gamma is an AI-powered presentation creation tool.

Main Feature: Gamma can create a complete slide deck from a simple paragraph or topic prompt.

How it works

Step 1: Enter a topic or paragraph.

Example prompt:

"Create a presentation about Artificial Intelligence for first-year engineering students. Explain AI, its applications, advantages, limitations, and future scope."

Step 2: Gamma generates a presentation structure.

Step 3: It creates slides with:

• Headings
• Text
• Images
• Layouts
• Design elements

Step 4: The student can edit and customize the slides.

Advantages of Gamma

• Saves time
• Quickly creates a complete presentation
• Provides attractive layouts
• Useful for students who are new to presentation design
• Helps organize ideas into a logical slide sequence

Example

Topic: "Applications of AI"

Gamma may automatically create:

1. Introduction to AI
2. AI in Healthcare
3. AI in Education
4. AI in Transportation
5. AI in Banking
6. Advantages of AI
7. Challenges of AI
8. Conclusion

Microsoft Copilot in PowerPoint

Microsoft Copilot brings AI capabilities directly into PowerPoint.

It can help students create and improve presentations without leaving PowerPoint.

Main Features:

A. Slide Generation

Copilot can generate slides based on a prompt or existing content.

Example:

"Create three slides explaining machine learning for beginners."

Copilot can generate appropriate slide content.

B. Design Suggestions

Copilot can help improve the appearance and organization of slides.

It can suggest:

• Layouts
• Visual arrangements
• Images
• Formatting
• Better presentation structure

This helps students create presentations that are easier to understand.

C. Speaker Notes

Copilot can help generate speaker notes for individual slides.

For example, a slide may contain:

Slide:

What is Artificial Intelligence?

• AI enables machines to perform tasks that normally require human intelligence.
• Examples include speech recognition and recommendation systems.

Speaker notes:

"Artificial Intelligence is a technology that enables computers to perform tasks that normally require human intelligence. Examples include recognizing speech, understanding images, and recommending content."

Canva AI Magic Design, text-to-presentation, brand kits; Beautiful.ai smart slide layouts

This guide provides a step-by-step workflow for creating high-impact academic presentations using modern AI design tools.

Part 1: Canva Magic Studio & Brand Kits

Step 1: Initialize the Brand Kit

Action: Go to "Brand" >"Brand Kit" >"Add New".

Task: Upload your institution's logo.

Action:  Select "Brand Colors". Input the specific Hex codes for your college colors.

Why: This ensures every AI-generated slide instantly looks official and professional.

Step 2: Magic Design (Text-to-Presentation)

Action: On the home screen, click the search bar and select "Presentations".

The Prompt: Type a specific engineering topic, e.g., "Design of Reinforced Concrete Structures for Seismic Zones”.

Action: Press Enter and let Magic Design generate several style options.

Step 3: Customizing with Smart Tools

Action: Choose a draft and click "Customize this template".

Branding: Go to the "Brand" tab in the editor and click your color palette. Select "Apply to all pages".

Part 2: Beautiful.ai Smart Slide Layouts

Step 4: Accessing Smart Slides

Action: Open Beautiful.ai. Click "Create New Presentation".

Difference: Unlike Canva, Beautiful.ai uses "Smart Slides" that automatically adjust layout as you add content.

Step 5: Choosing a Smart Template

Action: Click "Add Slide" and browse the template library (e.g., Timelines, Comparison Tables, Data Charts).

Task: Select a "Process" or "Cycle" diagram for engineering workflows.

Step 6: Dynamic Content Adjustment

Action: Start typing in a list or adding an image.

Observation: Notice how the other elements automatically resize and realign to maintain perfect white space.

Step 7: AI Rewrite (Designer Bot)

Action: Click the >"Designer Bot">icon.

Task: Ask it to "Summarize this technical paragraph into 3 bullet points."

Part 3: Advanced AI Workflows

Step 8: Magic Media - Text to Video

Action: In Canva, go to "Magic Media" > "Video".

The Prompt: Describe a moving scene, e.g., "Aerial view of a sustainable smart city with moving traffic and green spaces."

Why: Perfect for intros to catch students' attention.

Step 9: Magic Switch - Cross-Platform Conversion

Action: Click "Magic Switch" in the top bar.

Options: Select "Transform into Doc" to turn your slides into a detailed study summary or lab report instantly.

Step 10: AI Voiceovers (Murf.ai Integration)

Action: Go to Canva "Apps" and search for "Murf.ai".

Task: Paste your slide notes and select a professional voice.

Result: Converts a static presentation into an asynchronous lecture video.

 

 

 

 

Part 4: Comparison

 

Feature	Canva Magic Studio	Beautiful.ai
Best For	Visual flair and custom graphics	Data-heavy, structured layouts
AI Strength	Image generation & animations	Automatic layout & alignment
Branding	Manual Brand Kit application	Global theme settings

Comparing AI-generated vs. manually created presentations.

AI-Generated Presentation

An AI-generated presentation is created with the help of AI tools such as Gamma or Copilot.

Process:

Topic/Prompt → AI generates slides → Student edits → Final presentation

Advantages

·         Saves time

·         Quickly creates a complete slide structure

·         Provides design and layout suggestions

·         Helps students who are beginners in presentation design

·         Can generate speaker notes and supporting content

·         Useful when a presentation has to be prepared quickly

Limitations

·         AI may generate incorrect or outdated information

·         Content may be too general

·         Design may not exactly match the student's requirements

·         Students may become dependent on AI

·         The generated presentation still needs human checking and editing

Manually Created Presentation

A manually created presentation is prepared by the student using tools such as PowerPoint, Google Slides, or similar software.

Process:

Research → Plan → Write content → Design slides → Add visuals → Review → Present

Advantages

·         Complete control over content and design

·         Allows personal creativity

·         Content can be customized for a specific audience

·         Helps students understand the subject while preparing

·         Easier to develop a unique presentation style

Limitations

·         Takes more time

·         Requires presentation-design skills

·         Finding and organizing information can be difficult

·         Beginners may struggle with slide layouts

·         Creating professional-looking slides can require considerable effort

Aspect	AI-Generated	Manually Created
Preparation time	Usually less	Usually more
Content generation	AI-assisted	Student-created
Design	AI suggestions/automatic	Student-controlled
Creativity	AI-assisted	Strong personal control
Customization	Requires editing	High
Accuracy	Must be verified	Depends on research
Learning involvement	Can be lower if copied	Usually higher
Beginner friendly	Very useful	Requires practice
Human control	Moderate to high after editing	Very high
Best approach	AI + human review	Human research + design

 

Example

Topic: Artificial Intelligence in Education

Using AI

A student enters:

"Create a 7-slide presentation about the applications, benefits, challenges, and future of AI in education."

The AI may produce the initial slides automatically.

Manual Method

The student:

1.      Searches for information.

2.      Selects important points.

3.      Creates the slide structure.

4.      Writes the content.

5.      Finds images.

6.      Designs each slide.

7.      Reviews the presentation.

AI for document writing: Microsoft Copilot in Word — draft, rewrite, summarize;

What is AI for Document Writing?

AI for document writing means using Artificial Intelligence to help create, improve, and understand documents.

Microsoft Copilot in Word can assist users with tasks such as drafting, rewriting, summarizing, and improving documents.

Microsoft Copilot in Word

Microsoft Copilot is an AI assistant integrated into Microsoft Word. It can help users work with documents more efficiently.

Main Features

A. Draft

Copilot can create a first draft from a simple prompt.

Example:

"Write a one-page introduction about Artificial Intelligence for first-year engineering students."

Copilot can generate an initial document that the student can review and edit.

B. Rewrite

Copilot can help rewrite existing text.

It can be used to:

• Make sentences clearer
• Improve grammar and wording
• Change the tone
• Make text more concise
• Expand a short explanation

Example:

Original: "AI is useful in many areas and it is becoming important."

Improved: "Artificial Intelligence is increasingly being used across many fields, including education, healthcare, finance, and transportation."

C. Summarize

Copilot can summarize a long document and identify its main ideas.

For example, a 10-page report can be reduced to a short summary containing:

• Main topic
• Important points
• Key findings
• Conclusions

This is particularly useful when students need to understand lengthy documents quickly.

NOTE: Students should not blindly copy AI-generated content.

AI-generated text may contain:

·         Incorrect information

·         Missing context

·         Factual errors

·         Unnecessary or repetitive content

Gemini in Google Docs — writing assistance and inline Q&A;

What is Gemini in Google Docs?

Gemini in Google Docs is an AI writing assistant that helps users create and improve documents directly within Google Docs.

It can assist with writing, rewriting, summarizing, brainstorming, and answering questions about document content.

Writing Assistance: Gemini can help students create content from a simple instruction or prompt.

Example

A student can ask: "Write an introduction to Artificial Intelligence for first-year engineering students."

Gemini can generate a starting draft that the student can then review and modify.

It can also help with:

• Expanding ideas
• Improving sentences
• Changing the writing style
• Making text shorter or clearer
• Brainstorming ideas

Inline Q&A:

One useful feature is the ability to ask questions while working with document content.

Students can use Gemini to ask questions about their document or request changes to selected text.

Example:

Suppose a student has written: "Artificial Intelligence is used in healthcare, education, transportation, and banking."

The student can ask Gemini:

"Explain this paragraph in simpler language."

Or

"Give me two examples for each application."

Gemini can provide assistance without requiring the student to leave the document.

Notion AI — meeting notes, project plans, knowledge base;

What is Notion AI?

Notion AI is an AI assistant built into Notion, a workspace used for notes, project management, documentation, and knowledge sharing.

It can help students and teams organize information and turn notes into useful documents.

Meeting Notes: Notion AI can help organize and summarize meeting information.

For example, after a project discussion, students can turn their notes into:

• Main discussion points
• Important decisions
• Action items
• Tasks assigned to team members
• Follow-up points

Example

 

Raw notes:

"Discussed project topic. Ravi will collect datasets. Anu will prepare literature review. Team will meet on Friday."

AI can organize this into:

Project Tasks

• Ravi → Collect dataset
• Anu → Prepare literature review
• Team → Meet on Friday

Project Plans

Notion AI can help convert an idea into a structured project plan.

Example Prompt

"Create a project plan for a first-year engineering mini-project on smart agriculture."

It can help organize:

1. Project objective
2. Tasks
3. Timeline
4. Team responsibilities
5. Resources required
6. Expected outcome

This makes it easier for students to manage group projects.

Claude Projects — organizing documents and having a persistent AI assistant over a set of files.

What are Claude Projects?

Claude Projects are workspaces in Claude where users can organize a collection of documents and use Claude as an AI assistant that works with the information provided in that project.

Instead of repeatedly uploading the same files, students can keep relevant materials together and ask questions about them.

Organizing Documents

Students can add relevant files to a project, such as:

• PDF notes
• Text documents
• Reports
• Research papers
• Course materials
• Project documents

This creates a central knowledge area for a particular subject or project.

Example: A student creates a project called:

"Data Science Semester Project"

They can organize materials such as:

📄 Project proposal
📄 Research papers
📄 Dataset documentation
📄 Course notes
📄 Project report

Persistent AI Assistant

Claude can use the information in the project as context when responding to questions.

For example, after adding course notes, a student can ask:

"Explain the concept of supervised learning using my uploaded notes."

Or:

"Summarize the important points from these research papers."

The student can continue working within the same project instead of starting from scratch each time.

Useful Applications

📚 Studying

Ask questions about uploaded class materials.

🔬 Research

Organize research papers and compare their key ideas.

💻 Projects

Keep project documents together and ask Claude for assistance.

📝 Report Writing

Use project materials to help structure and improve reports.

📖 Revision

Create summaries, questions, and explanations from study materials.

AI Integration

Use Gamma.app to generate a presentation on an engineering topic in under 2 minutes; use Copilot in Word or Gemini in Docs to draft and improve a report; compare ChatGPT, Claude, Gemini, and Microsoft Copilot on the same writing task; use Grammarly to review your own writing.

 AI POWER-UP CHALLENGE

Can AI Make You a Better Engineer?

 

Objective: Use different AI tools to create, improve, compare, and evaluate engineering-related content.

Team: 3–4 students per team.

Time: 60–90 minutes.

Mission 1: 2-Minute Presentation Challenge

Choose an engineering topic and use Gamma.app to create a 5–7 slide presentation.

Record:

·         Topic

·         Prompt used

·         Time taken

·         Number of slides generated

·         Changes made by your team

Mission 2: AI Report Builder

Use either Microsoft Copilot in Word or Gemini in Google Docs.

Create a 500-word report on your selected engineering topic.

Ask the AI to:

1.      Draft the report.

2.      Rewrite one paragraph.

3.      Simplify technical language.

4.      Summarize the report.

Edit the final report yourself.

 

Mission 3: AI Face-Off

Give the SAME prompt to:

·         ChatGPT

·         Claude

·         Gemini

·         Microsoft Copilot

Compare the outputs based on:

·         Clarity

·         Accuracy

·         Organization

·         Examples

·         Language

·         Usefulness

Do not simply copy the AI outputs. Explain your observations.

Mission 4: Beat the AI

Write one paragraph yourself without AI.

Use Grammarly to review it.

Record:

·         Problems identified

·         Corrections suggested

·         Corrections accepted

·         Corrections rejected

Explain whether Grammarly improved your writing.

Final Reflection

Answer:

1.      What did you learn about AI tools?

2.      What differences did you observe between the AI systems?

3.      What mistakes did you find in AI-generated content?

4.      What changes did you make yourself?

5.      Why is human judgment still important?

Final Presentation

Prepare a 3-minute presentation covering:

·         Your engineering topic

·         AI tools used

·         Gamma challenge

·         Report-writing challenge

·         AI comparison

·         Grammarly review

·         What your team learned

Golden Rule:

ASK AI → CHECK AI → IMPROVE AI

AI is your assistant.
You are the engineer.

UNIT-I Question:

1.       What is Artificial Intelligence (AI)? Explain the differences between AI, Machine Learning (ML), Deep Learning, and Generative AI with suitable examples.

2.      Trace the brief history of Artificial Intelligence from the Turing Test to ChatGPT and the current Generative AI era. Mention the major developments and milestones.

3.      Explain how Large Language Models (LLMs) work. Discuss:

Training data

Next-token prediction

Temperature

Context windows

Knowledge cutoffs

Hallucinations

4.      Compare ChatGPT, Claude, Gemini, and Microsoft Copilot. Explain their major strengths, differences, and suitable use cases.

5.      Explain how AI can be used for creating presentations. Compare the features and uses of:

Gamma.app

Microsoft Copilot in PowerPoint

Canva AI

Beautiful.ai

6.      Explain the use of AI tools for document writing and productivity.

7.      What are the principles of responsible and ethical use of Generative AI in academic work?

8.      Discuss AI-generated content and academic integrity.

 

 

 

 

 

********** END OF UNIT-I **********