Chatbot

Custom AI chatbot implementation framework

Supplier landscape

Other information

Introductory presentation incl. use cases

Custom AI Chatbots

Custom AI Chatbots

Introduction, Use cases, Getting started

Christian Ullrich

Outline

  1. Introduction
  2. Implementation
  3. Getting started
  4. Use cases for custom AI chatbots
  5. See it in action
  6. Closing remarks

1. Introduction

What is artificial intelligence?

  1. AI is about making computers behave intelligently.
  2. Some keywords: Machine learning, neural networks, natural language processing
  3. AI performs tasks that need human intelligence: More than logical rules.

What is generative artificial intelligence?

  1. Generative AI creates new content: Writing, images, and melodies.
  2. Realistic, coherent outputs.
  3. ChatGPT is an example of generative AI: Generating relevant sentences based on input.

What is ChatGPT?

  1. ChatGPT is an AI software for human-like text based on input.
  2. Useful for brainstorming, document outlining, and content drafting.
  3. Not so useful for calculations, facts, and research.

How can we use ChatGPT?

  1. ChatGPT can draft emails, write code, write documents, and translate languages.
  2. Many tasks involving text generation.
  3. Challenge: Find your individual use cases.

What are information repositories?

  1. Information repositories store and maintain (written) information.
  2. Websites, internal wikis, knowledge base, research databases
  3. A broad term for many existing things.

Why do we need more public information repositories?

  1. Accessibility, transparency, accountability
  2. Making information available to anyone: Democratizing information.
  3. Political strategy: Data legislation, Freedom of information acts

How can we create information with the help of ChatGPT?

  1. ChatGPT reduces the amount of work to create high-quality written information.
  2. ChatGPT can rework existing low-quality content.
  3. ChatGPT can simplify complex and complicated content.

What are classical chatbots?

  1. Classical chatbots simulate conversations with humans based on predefined rules.
  2. They struggle to handle unexpected user inputs.
  3. Classical use case: Customer service to handle simple queries

What are custom AI chatbots?

  1. Internal information + ChatGPT capability.
  2. Specific sources (+ ChatGPT, if desired).
  3. New software category after ChatGPT was launched.

How can we use custom AI chatbots?

  1. Customer service, marketing, sales. And much more.
  2. Answer questions. Research trusted sources. Act as a 24/7 advisor.
  3. Like a classical chatbot but better, faster, (cheaper).

2. Implementation

How can we use ChatGPT for our organization?

  1. Research
  2. Brainstorming
  3. Creation
  4. Rework
  5. Preparation

How can we share more information?

  1. Change your mindset.
  2. Create more.
  3. Share more.

How can we use a custom AI chatbot?

  1. Search on public websites.
  2. Search through semi-public content. (e.g., magazines)
  3. Internal search. (Data privacy challenge)
  4. Build a research alliance.

How can we differentiate with a custom AI chatbot?

  1. Faster, better, cheaper information dissemination.
  2. New business models.
  3. Public relations. (”We use AI”)
  4. Learning.

3. Getting started

Try out ChatGPT!

  1. Experiment with ChatGPT.
  2. Try different prompts to understand its capabilities.
  3. Learn “prompt engineering”.

Share more information publicly!

  1. Identify information that can be shared publicly.
  2. Align your internal stakeholders. (Aaargh!)
  3. Share it.

Try a custom AI chatbot software tool!

  1. Set up a project.
  2. Align your stakeholders. (Aaargh!)
  3. Set up the technology in a day. (Literally!)
  4. Follow a methodology.

Work together!

  1. Try ChatGPT indivudally.
  2. Set up a “community of practice”.
  3. Share your experiences within the organization.

Get help!

  1. Don’t outsource.
  2. Build your methodology with the help of ChatGPT.
  3. Get a trusted advisor.

4. Use cases for custom AI chatbots

Generic use cases

Marketing & Sales

  1. Website search
  2. Customer education
  3. Information consolidation

Customer service

  1. Information already exists.
  2. Less-prepared internal content.
  3. 24/7 customer agent. (But actually working.)

Internal operations

  1. Brainstorming and writing
  2. Search & Advice
  3. Process execution based on natural language

Research available to the public

  1. Search at your organizational website.
  2. Research on websites of various organizations.
  3. Separate business models.

Information-rich service providers

  1. All professional services
  2. Digital business development
  3. Separate business models

Armed Forces

Training

  1. Share training material more openly (especially if created with ChatGPT).
  2. Offer trustworthy(!) information repositories.
  3. Make training material adaptable (but still correct and trustworthy).

Standard Operating Procedures

  1. Write additional practices and procedures with ChatGPT.
  2. Share them publicly.
  3. Create an AI showcase project.

Recruitment

  1. Offer a 24/7 recruitment office. (Working business case!)
  2. Offer educational information for high school students. (e.g., computer science courses) (role model: America's Army)
  3. Offer 24/7 online teaching on relevant topics.

Public relations

  1. Collect information repositories (and there are many).
  2. Make information “accessible” (and even more accessible).
  3. Last but not least: Showcase AI adoption to stakeholders (e.g., journalists).

Defense research

  1. Collect information repositories across organizational boundaries.
  2. Align stakeholders (Aargh!) across organizational boundaries.
  3. Showcase AI adoption in research.

Chambers of Commerce

a. Website search

  1. Website search is nearly always awful if it’s not based on Google.
  2. Custom AI chatbot as an addition, not a replacement.
  3. Interactive guidance instead of finding the right page.

b. Training & Education

  1. Educational content hub
  2. 24/7 teacher based on trusted sources.
  3. New business models together with other organizations.

c. Economic research

  1. Explanation instead of search.
  2. Complex content for new target groups.
  3. Research for practical guidance.

d. Legislative advocacy (aka Lobbying)

  1. Explanation instead of search.
  2. Explanation for more target groups.
  3. Joint lobbying with other organizations.

e. Business directories

  1. Business listings 2.0
  2. Find the right business without maintaining a directory.
  3. Make trustworthy educational information accessible.

Education

AI learning

  1. Courses about AI software tools and their use.
  2. Practical adoption of software tools.
  3. Learning the value of information sharing.

Teaching

  1. Trustworthy learning content.
  2. 24/7 support.
  3. AI showcase implementation.

Student content sharing

  1. A platform for sharing information created by students (with or without the help of ChatGPT).
  2. Re-using already created content by other classes.
  3. Teaching the sharing economy.

Inter-organizational information exchange

  1. Making information accessible without publishing them in one information repository.
  2. Building networks of trustworthy sources. (even competing views)
  3. Limited organizational alignment, limited “politics”.

New business models

  1. New business models based on open-source information.
  2. Offering information with self-service customization.
  3. Making already free information accessible.

IT management

a. Digital strategy

  1. Make the strategy “useful” through AI guidance.
  2. Discuss strategies openly.
  3. Showcase your AI implementation.

b. Standard operating procedures

  1. Make SOPs practical.
  2. Share SOPs openly.
  3. Develop a SOP platform.

c. User help & training

  1. Make guides practical.
  2. Offer 24/7 support.
  3. Make training material applicable.

d. Service catalog

  1. Publish public service catalog as AI showcase.
  2. Offer intelligent search.
  3. Brainstorm procurement business models.

e. Sourcing & Supplier management

  1. Improve contract (lifecycle) management.
  2. Share contract templates with the public.
  3. Help with contract writing based on trustworthy sources.

Health

Citizen information

  1. Find answers from a trustworthy source.
  2. Combine several trustworthy sources into a knowledge hub.
  3. 24/7 virtual paramedic.

Professional research

  1. Q&A for professionals.
  2. Education for professionals.
  3. Scientific research.

Policy & Regulation

  1. Citizens
  2. Citizen advisors
  3. User help desk

New business models

  1. Information dissemination
  2. Shift from medic to paramedic.
  3. Disruptive business models in emerging countries.

Inter-government cooperation

  1. Platforms for publicly funded information.
  2. Integrated translation.
  3. Public health policies & practices exchange without document exchange or translations.

Management frameworks

Practicability

  1. Applicable management information.
  2. Open frameworks instead of bullshit bingo books.
  3. Concise answers instead of 230 pages.

New frameworks

  1. Lean testing of new management approaches.
  2. No book, no training, no implementation.
  3. Build, share, wait.

Market demand

  1. Nothing new but constant renewal of existing concepts.
  2. New focus: Simplification & Increase of speed
  3. New target users: Students, Frontline workers, Specific professions

Human experiences

  1. Collection of actual experiences.
  2. Reports as basis for answers.
  3. No need to read long interviews.

New approaches

  1. New approaches to “framework writing”.
  2. Significantly faster development with ChatGPT.
  3. New business models. (Content can/will/must be free.)

Political communication

Self-service information

  1. Stakeholders pull information themselves.
  2. Stakeholders know how to use information.
  3. Stakeholders need less manual customization.

New target groups

  1. Reaching new target groups with less effort.
  2. Less information customization.
  3. Quicker target group trials.

Inter-organizational cooperation

  1. Cooperation without a central information repository.
  2. Simple onboarding and offboarding of information repositories.
  3. Chatbot links sources: Website traffic generator.

Internationalization

  1. International cooperation with less effort (e.g., translations).
  2. Re-use of information resources (and even repositories).
  3. Collaboration on information development (with future customization).

New business models

  1. Private/Public funding sources allow new business model testing.
  2. Open work environment between organizations facilitates ideation.
  3. Evident need for efficiency gains in policy development.

Political education

Perfect use case

  1. Information shall be distributed openly and freely.
  2. Nearly no commercial or other restrictions.
  3. Inter-organizational cooperation is already happening.

Organizational cooperations

  1. Making information accessible without publishing them on one platform.
  2. Building networks of trustworthy sources. (even competing views)
  3. Limited organizational alignment, limited “politics”.

Self-service information

  1. Less teaching, more self-service learning.
  2. Less central customization necessary.
  3. Scalability of the political education model.

Simpler customization

  1. “Everyone can be a teacher.”
  2. AI-assisted self-service customization.
  3. Platform instead of pipe business model possible.

Improved accessibility

  1. Content is actually used.
  2. Possibly less but more intense use.
  3. Content use by more target groups.

Procurement

Process design

  1. Make your purchasing guideline accessible to suppliers.
  2. Publish process descriptions and SOPs.
  3. Develop industry standards.

Contract (lifecycle) management

  1. Make contracts accessible.
  2. Offer a 24/7 contract management office.
  3. Answer questions by using an AI chatbot but not making the chatbot accessible.

Statement of work

  1. Share your statements of work with other organizations.
  2. Offer a trustworthy(!) statement of work assistant.
  3. Integrate a chatbot for suppliers into your procurement process.

Supplier directory

  1. Crawl suppliers’ websites.
  2. Offer a supplier research chatbot.
  3. Develop a business model.

Advanced process support

  1. Integrate your internal apps with classical and custom AI chatbots.
  2. Execute processes with natural language.
  3. Improve processes based on prompt analysis.

5. See it in action

ChatGPT

  1. Try it yourself.
  2. Conduct a workshop.
  3. Share your experiences.

Information repositories

  1. Websites
  2. Scattered research
  3. Internal information to be published

Custom AI chatbot

  1. customgpt.com
  2. intrenion.com
  3. TBD…

Workshops

  1. ChatGPT
  2. Information management
  3. Custom AI chatbots

6. Closing remarks

Start simple!

  1. Try ChatGPT.
  2. Share your experience.
  3. Share your results.

Think from the goal!

  1. Where would a custom AI chatbot be useful?
  2. What information repository do we need for that?
  3. How can I start with ChatGPT?

Why generative AI?

  1. It works.
  2. It will change everything.
  3. It works for you. (After max. 30 min of brainstorming.)

Why now?

  1. It’s useful now.
  2. It’s accessible now.
  3. It’s affordable now.

Why you?

  1. Someone has to do it.
  2. Lead, follow, or get out of the way. (General Patton)
  3. Grow yourself, grow your career, grow your organization, grow humanity.

Why together?

  1. Think collaboratively. (It’s hard, I know.)
  2. Join forces.
  3. Inspire others.

Last remarks

  1. My humble opinion.
  2. Discussion is just starting.
  3. Other opinions are welcome.

Questions & Discussion

Christian Ullrich

intrenion.com

linkedin.com/christianullrich