0% found this document useful (0 votes)
7 views33 pages

Lecture 2

The document discusses the use of Generative AI in business communication, highlighting its capabilities in content development, data analysis, and personalization, while also addressing the operational, ethical, and legal risks associated with AI usage. It emphasizes the importance of ethical communication, transparency, and the need for human skills in an AI-driven environment. Additionally, it outlines the mechanics of unethical communication and strategic ethical decision-making in business contexts.

Uploaded by

bn9248
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
7 views33 pages

Lecture 2

The document discusses the use of Generative AI in business communication, highlighting its capabilities in content development, data analysis, and personalization, while also addressing the operational, ethical, and legal risks associated with AI usage. It emphasizes the importance of ethical communication, transparency, and the need for human skills in an AI-driven environment. Additionally, it outlines the mechanics of unethical communication and strategic ethical decision-making in business contexts.

Uploaded by

bn9248
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Lecture 2.

Artificial Intelligence &


Ethics in Business
Communication
Duy Khanh NGUYEN
Generative AI
Module A: Using AI in Business Communication
Part 1
How Generative AI Works
1.1 Defining Core Technologies

Artificial Intelligence (AI) Generative AI (GenAI) Large Language Models


(LLMs)
Technology simulating human A specific type of AI that can create new
intelligence. It mimics the result of data (text, code, AV) rather than just Statistical prediction engines. They do

thinking via algorithms; it does not categorizing existing data. not "know" strategy; they calculate

possess a brain or consciousness. probability. It is math, not magic.


1.2 Strategic Distinction: Search vs. GenAI

Search Engines (The Generative AI (The


Librarian) Author)

Role: Retrieval. Role: Composition.

Function: Finds existing Function: Combines


information and links to the patterns to write new content
source. word-by-word.

Use Case: Fact-finding, Use Case: Ideation, drafting,


citation sourcing, verification. summarizing, brainstorming.
Part 2
Strategic Applications in Business
2.1 Content Development & Knowledge Management

Creative Partner: GenAI helps solve "writer's block" by brainstorming ideas and making outlines.

Knowledge Management:

Problem: Information in companies is often separated and hard to find (silos).


Solution: AI can scan internal databases to find information or experts.

Example: EigenTechnologies uses natural language processing (NLP) to analyze thousands of complex contracts for
mergers and acquisitions (M&A) and legal compliance.
2.2 Data Analysis & Insights

Sentiment Analysis: AI "listens" to social media to measure public feeling about a


brand. This helps companies react quickly to crises.

Predictive Analytics: Forecasting customer behavior to choose the best messages.


2.3 Personalization & Efficiency

Mass Customization Efficiency Accessibility

Sephora uses AI to alter website layouts The Associated Press (AP) automates Zoom offers live captions and
in real-time based on user behavior. routine financial and sports reporting, translation, removing language barriers
freeing journalists for deep reporting. in global meetings.
2.4 Professional Development (The Coach)

Use AI as a coach to improve your skills, not just to do the


work for you.

The Editor: "Evaluate my writing based on 5 criteria: Clear,


Concise, Practical, Factual, and Persuasive."

The Opponent: "Act as a negotiation partner or a difficult


customer so you can practice handling situations."
Part 3
Risks, Ethics, and Governance
3.1 Operational Risks: Hallucinations & Reliance

AI Hallucinations Overreliance (Laziness)

AI can confidently invent information that is not true. If you let AI do all the thinking, you lose value.

Imagine this: Your CEO asks you about a strategy in


your report, and you cannot answer because AI
wrote it, not you.
3.2 Ethical & Social Risks

Deepfakes & Fraud: Technology that fakes audio and


video is a threat to financial security.

“An employee in Hong Kong was tricked into transferring


$25 million to criminals after joining a video call where the
CFO and colleagues were all fake.”

Bias: AI can make historical biases about race and


gender worse if the training data is biased.

Environmental Impact: AI uses a huge amount of


electricity and hardware resources. Companies need to
consider their "carbon footprint" when using AI.
3.3 Legal Risks & Transparency

Intellectual Property (IP):


Input: Legal arguments about using copyrighted
data to train AI.

Output: In the US, work created entirely by AI


cannot be copyrighted.

Data Privacy: Be careful with laws like GDPR (Europe)


or CCPA (California) when putting personal data into AI.

Disclosure (Transparency): In a professional setting,


you should be open about using AI. If your boss asks
you to write a report, they expect you to write it (or
carefully edit it), not just copy-paste it from a machine.
Part 4
Operational Mastery: Prompt Engineering
4.1 Three Basic Types of Prompts

Directive Question-based Creative

Tell it to do a specific task. Ask for an answer. Ask for new ideas.

Ex: "Summarize this article." Ex: "Why did this strategy fail?" Ex: "List 10 names for a new
product."
4.2 Interaction Strategy

Provide Context: Act like a manager giving a task to a new


employee. Clearly state: The Goal, The Audience, and The
Tone.

Iterative Process: Do not stop at the first result. Ask AI to


refine it: "Make the tone more neutral," or "Add specific
examples."

Check the Output: Always Fact-check. AI can invent citations


that look very real.
Part 5
Managing Tech & Human Skills
5.1 Fighting Information Overload

You must filter information for yourself and others. Do


not send meaningless messages. 'Prune' (cut back)
unnecessary information channels."
5.2 The Urgency of Human Skills

To avoid being replaced, focus on what machines cannot do:

Emotional Intelligence (EQ): Understanding hidden emotions.

Contextual Awareness: Understanding office politics and culture.

Reconnecting: When email causes confusion, break the loop. Pick up the phone or meet in person. No
technology can replace a sympathetic smile or a real handshake.
Ethical
Foundations
Module B: The Foundations of Ethical Communication
Part 1
The Foundations of Ethical Communication
1.1 Core Definitions

Ethics Ethical Communication

The accepted principles of right and wrong that Communication that includes all information an
govern behavior and decision-making within a audience needs to make an informed decision or take
society. an informed stance. It must be:

Complete: Lacks nothing required for


understanding.

True: Factually accurate.

Non-deceptive: Does not mislead, intentionally


or inadvertently.
1.2 The Transparency Mandate
The growth of social media and AI has elevated transparency to a
critical concern. Transparency is defined as a sense of
openness—giving all participants access to the information needed to
accurately process messages.

"Audiences deserve to know when they are being marketed to and who is
behind the message."
Part 2
The Typology of Unethical Communication
2.1 Mechanics of Deception

Withholding Information Stealth Marketing Native Advertising &


Intentionally omitting critical facts
Influencers
Communication activities that are
that prevent the recipient from marketing efforts in disguise. Sponsored content designed to

acting in an informed manner look like legitimate news.


(Product Placement, Guerrilla
(e.g., failing to disclose safety risks Influencers must disclose
Marketing)
or budget problems). relationships (#ad, #sponsored).
2.2 Distorting Information

Statistical Manipulation Visual Manipulation & Deepfakes

Using averages to conceal extreme outliers or Visual Manipulation: Altering photos or changing
manipulating trend calculations to suggest graph scales to exaggerate/conceal differences.
unsupported future outcomes.

Ex: Boasting of a 40% increase that only recovers Deepfakes: AI-generated audio/video content that

previous losses. pushes distortion to unprecedented levels.


2.3 Plagiarism and IP Violations

Plagiarism & Copyright Fair Use

Plagiarism: Presenting another's words, creative A legal defense allowing limited use of copyrighted
product, or ideas as one's own. material (often for non-commercial purposes).

Copyright: Legal protection for originators of creative Fair use is a defense invoked after you are sued. You
content. You can be sued for infringement even cannot preemptively label content as "fair use" to
without profiting. avoid liability.
2.4 Failing to Consider Negative Consequences

Even truthful communication is unethical if the sender fails to Case Study: Lush Cosmetics
anticipate harmful outcomes.
Lush Cosmetics decided to "unplug" from certain
social media platforms, equating them to a "dark and

Example: Promoting health advice suitable for professionals dangerous alleyway," because they believed the

to a general teenage audience without medical context. platforms were detrimental to users' mental health.
2.5 Violating Privacy

Data Aggregation
Data brokers collect and sell vast profiles on consumers
(health, income, habits) to insurers and marketers.

Workplace Surveillance
Employers monitoring keystrokes, location, and attention
raises severe ethical tensions regarding the invasion of
personal space vs. the protection of company
resources.
Part 3
Strategic Ethical Decision Making
3.1 Distinguishing Dilemmas from Lapses

Ethical Dilemma Ethical Lapse

A situation involving a choice between ambiguous A clearly unethical or illegal choice.


alternatives.

The Scenario: Competing valid interests. The Scenario: Deliberate dishonesty or illegality.

Example: Employees want higher wages (valid) vs. Example: Telling a client you can meet a deadline
Investors want lower costs for stock growth (valid). when you know you cannot. There is no dilemma
Both positions have ethical merit; the solution is in the here; it is simply wrong.
gray area.

"A true dilemma forces you to choose between two 'rights,' not between 'right' and 'convenient.'"
Questions?
Thank you for your time.

You might also like