What will we cover

• What is AI?
today?
• The AI landcsape

• Fears

• Why use AI?

• AI tools: categories

• What can it do for me?

The AI landscape
 Why use AI
1. Efficiency: They help automate repetitive tasks,
saving time and energy.

2. Accuracy: AI can analyze huge amounts of data

with fewer errors than humans.

3. Personalization: AI tools can tailor experiences

based on individual preferences, like recom-
mending music or products.

4. Accessibility: AI makes technology more

accessible, such as voice-to
text for those with disabilities.

5. Innovation: They open up new possibilities in

fields like medicine, art, and education,
pushing the boundaries of what's possible.
 T Y
C TIVI
OD U
D P R
A SE
RE
INC
 What
about you?

How do you use AI at work / home?

Most common reasons people fear AI

• Fear # 1: “AI will

replace my job.”
• Fear # 2: “AI creates
soulless work.”
• Fear # 3: “AI will make
everyone a creator.”
 Categories of AI tools
➡ Text Generators: These tools produce written content that is both fluent and
intelligible. They can generate essays, blogs, scripts, news articles, and even poetry

➡ Image Generators: These tools create visuals from text prompts, ranging from
photorealistic portraits to surreal landscapes. They can produce diverse images in a
range of mediums, from photorealistic oil painting style to anime.

➡ Audio Generators: These tools compose original music in a variety of styles, as well
as voices. AI music generators analyze music tracks and metadata to identify patterns
and features in particular music genres, and have also been trained on lyrics
associated with songs.

➡ Video Generators: These tools produce unique video clips from scratch when a user
inputs a text prompt. Some programs have harvested existing videos to learn how to
create new ones, while others source the audio, visual, and text elements to create
videos.
 Categories of AI tools
‣ Code Generators: These tools automatically write their own code, fix bugs in existing code,
and translate between programming languages. Generative AI is exposed to large datasets
of open access code in a variety of program languages and can find common patterns,
practices, and structures within program languages. Examples include ChatGPT, CodeT5,
and Tabnine.

‣ AI note takers: are AI-powered tools designed to generate accurate meeting transcripts and
summaries, allowing users to focus on discussions without missing important information.
These tools can capture real-time meeting transcripts, generate summaries, and identify key
insights.

‣ AI music generators: are software tools that use artificial intelligence to create music based
on user inputs or pre-defined algorithms. These generators analyze extensive datasets of
music, learning from a variety of styles and compositions. AI music generators have grown
into different categories for musicians and non-musicians alike.
 Categories of AI tools

‣ AI email assistants: An AI email assistant is a software tool or digital helper that uses artificial
intelligence to streamline and improve email management. It helps automate mundane
tasks, improve efficiency, and provide insights into email content and interactions

‣ AI workflows: AI workflows are structured sequences of steps that use artificial intelligence
to complete tasks, automate business processes, and optimize operations. They leverage AI
to ensure tasks are completed efficiently and accurately . AI workflows streamline, optimize,
3

and often completely transform business processes . 6

 Various functions for hospitality.
Learning and Content creation &
Creativity
development editing

Social media
Logistics HR
management

Customer service Writing Marketing

Administration Business strategy Finance

Legal
 TOP TOOLS
Text Generators

• OpenAI GPT-5: Known for its advanced text generation capabilities, GPT-5
can write content, assist with programming, and provide live chatbot support.
It supports over 50 languages and offers both free and premium plans.

• Claude: Known for its writing capabilities

• Copilot for 365: part of the Microsoft offering and availbale to all 365 users.

• CopyAI: Copy.ai generates persuasive copy for ads, emails, and social media
posts. It supports multiple languages and offers a range of templates.

• Jasper AI: Focused on long-form content creation and marketing

• My favourites: Perplexity AI / Google AI studio ( Gemini )

 TOP TOOLS
Image Generators
• DALL-E 3: Generates images from textual descriptions. ( Part of OpenAI)
• Midjourney: Popular for high-quality image outputs. Only offers paid options.
• Stable Diffusion: Best for technically inclined users, developers, those wanting
maximum control, customization, specific styles via fine-tuned models, and
potentially free local generation. Also video, audio and 3D. ( stability.ai )
Requires technical setup for local use, or via various web platforms.
• Adobe Firefly: Designed with creative professionals in mind. Access via Adobe
Creative Cloud apps, Adobe Express, and the Firefly web interface.
• Leonardo.Ai: A user-friendly platform built heavily on Stable Diffusion but also
offering its own fine-tuned models. Has a free option.
• My favourite: Ideogram
 Please create an image of a sixty year old bald man,
running a marathon.
He is wearing a light blue running vest and black shorts.
Along the road people are cheering him on.
He is smiling as he runs passed them.
 TOP TOOLS

Video Generators

• Synthesia: Allows users to create videos using AI avatars.

• Runway: Offers creative video generation tools.

• Invideo, Descript

• My favourites: Puppetry & OpusClip

 TOP TOOLS

Audio Generators

• ElevenLabs: Provides a library for realistic AI voice

generation.

• Suno: Focuses on music generation.

• Imagine Art: music.

• My Favourite: ElevenLabs and Suno

IMAGINE.ART
 TOP TOOLS
Code Generators

• GitHub Copilot: Assists in writing code by providing suggestions

based on context.

Additional Tools

• Magic Write: Integrated within Canva for brainstorming and

content generation.

• Notion AI: Helps in organizing notes and enhancing productivity.

• My favourite: Turboscribe, Fireflies

 Online
meeting note
Takers
When you write a prompt, you are attempting to set up the LLM
to predict the right sequence of tokens. Prompt engineering is
the process of designing high-quality prompts that guide LLMs
to produce accurate outputs. This process involves tinkering to
find the best prompt, optimizing prompt length, and evaluating a
prompt’s writing style and structure in relation to the task. In the
context of natural language processing and LLMs, a prompt is an
input provided to the model to generate a response or
prediction.

Prompt engineering
General prompting / One-shot & System

zero shot few-shot prompting

Contextual Role Step-back

prompting prompting prompting

Chain of Self-consistency

Thoughts (CoT) prompting

Prompting techniques
 Code
Tree of Thoughts
prompting
(ToT)

ReAct

(reason & act)

Automatic Prompt

Engineering

Prompting techniques
When you choose examples for your prompt, use examples that are
I would like 2 suggestions on how to approach a
relevant to the task you want to perform. The examples should be
problem
diverse, I have
of high withand
quality, a friend:
well written. One small mistake can
Example
confuse the 1: Youand
model havewill upset
result inme and I output.
undesired need you to
know
If how
you are I feel.
trying to generate output that is robust to a variety of
Example
inputs, then 2:
it isCan you please
important explain
to include to me
edge cases whyexamples.
in your you
Edge cases you
did what are inputs thatme?
did to are unusual or unexpected, but that the
model should still be able to handle.

One-shot &

few-shot

Prompting techniques
Prompt
Context: You are writing for a blog about retro 80's arcade video games.
Suggest 3 topics to write an article about with a few lines of description of what this
article should contain. By providing contextual prompts, you
Output can help ensure that your AI
interactions are as seamless and
1.The Evolution of Arcade Cabinet Design:
This article would explore the evolution of arcade
efficient as cabinet designs,
possible. The from the will
model earlybe
wood
Contextual
and metal cabinets of the 1970s to the sleek, neon-lit designs of the 1980s and
able to more quickly understand your
beyond.
2.Blast prompting
request and be able to generate more
From The Past: Iconic Arcade Games of The 80’s:
- This article would feature a glimpseaccurate and relevant
into the history, responses
impact, and .
legacy of some of
the most iconic and influential arcade games of the 1980s, highlighting their
innovations, popular mechanics, and enduring charm.
3.The Rise and Retro Revival of Pixel Art:
- This article would delve into the evolution of pixel art as a defining visual style of
early arcade games, tracing its roots, exploring its techniques, and discussing the
resurgence of pixel art in modern games and digital art.

Prompting techniques
Prompt

I want you to act as a travel guide. I will write to you about my location and you will suggest 3 places
to visit near me. In some cases, I will also give you the type of places I will visit.
My suggestion: "I am in Amsterdam andRole I wantprompting is a technique in
to visit only museums."
prompt engineering that involves
assigning a specific role to the gen AI
Output
Role model. This can help the model to
1.Rijksmuseum: World-renowned museum, known for Dutch masterpieces like Rembrandt’s “The
Night Watch.” Don’t miss the incrediblegenerate
collection ofmore relevant
paintings andshowcasing
and artifacts informative
Dutch art
prompting
and history. output, as the model can craft its
2.Van Gogh Museum: Immerse yourself responses
in the captivating worksspecific
to the of Vincentrole
van Gogh.
that The museum
it has
houses the largest collection of his paintings and drawings, including “The Starry Night” and
“Sunflowers.” been assigned.
3.Stedelijk Museum Amsterdam: Discover modern and contemporary art from around the world. The
museum’s collection includes works by Picasso, Kandinsky, and Rietveld, housed in a striking
modern building.

Prompting techniques
Automatic Prompt

Engineering
 Best practice
➡ Provide examples
➡ Design with simplicity
➡ Be specific about the output
➡ Use Instructions over Constraints
➡ Control the max token length
➡ Use variables in prompts
➡ Experiment with input formats and writing styles
➡ Adapt to model updates
➡ Experiment with output formats
➡ Experiment together with others
https://huggingface.co/spaces
Still overwhelmed?
Where to start?
Which tools to
use?

YOU NEED AN AI
 Key Points
✴ Experiment

✴ Check / Verify

✴ Keep learning

✴ Don’t abuse

✴ Stay original

✴ Create your “team”

ENTERPRISE
SOLUTIONS
 AI STRATEGY
1. Clear Goals
2. Data Rules
3. Solid Tech Foundation
4. Employee Buy-In
5. Ethics Guidelines
6. Organizational Readiness

The key is making an AI strategy straightforward - define goals, get

the right data/tech foundations, educate employees, ensure ethical
practices, and get the organization ready for AI-driven changes.
 AI POLICY
1. Clear rules for AI
2. No sneaky AI
3. Protecting your info
4. Fairness for all
5. Humans in charge
6. Using AI for good
7. Keeping Up and Learning
8. Keeping AI policies up-to-date
AI agents are software systems that use
artificial intelligence to autonomously
perform tasks on behalf of users or
other systems. They can set goals,
gather information, plan actions, and
learn from experience, essentially
acting as digital assistants that can
handle complex tasks.
[email protected]