AI is Not Just a Tool: It’s a Language for Thinking and Communication. Four Essential Skills for AI Literacy in the Age of AI
Being Human for a Better Tomorrow in the Age of AI
In the last 3 months, I have progressed from a casual observer to an active user of Gen AI for research, learning and business solution thinking. As an AI technophile, I am also reading and engaging about the technology to see what is under hood, to grasp the capabilities, limitations and governance for beneficial use of AI.
Using LLM tools as a thought partner, reading, listening, exploring the technology with experts has been exhilarating like the joys of learning in school days without burdens and boundaries. I have also been writing and pushing that everyone must and indeed can be AI literate.
Learning AI is less about technology know-how (of course does not apply to SDEs!) and closer to how learning a language and how that changes your mode of thinking. The Sapir-Whorf Hypothesis (linguistic relativity) suggests that the structure of a language affects its speakers’ cognition and worldview.
Just as language proficiency involves mastering different skills like speaking, listening, reading, and writing, AI proficiency requires building skills in thinking about applications, questioning, reasoning and making judgements to discriminate between true and false.
These four essential skills, illustrated with simple examples, are essential for anyone looking to effectively use AI tools, much like language skills are crucial for communication.
1. Application Thinking: Being Curious, Exploring Use Cases Where AI can be helpful.
Being Curious, Exploring Use Cases Where AI Can Be Helpful
Application thinking is about identifying where AI can add value, whether through curious exploration, task orientation, or finding novel solutions to unsolved problems.
Exploratory Example:
Imagine you’re a sci-fi enthusiast looking for movie/web series recommendations. You could ask AI for recommendations, specifying likes like ‘Odyssey 2001’, ‘Blade Runner’ or ‘The Three-Body Problem’ and those that don’t align with your tastes, like say ‘District 9’.
Task-Oriented Example:
Reviewing a job description (JD) for completeness and impact. AI can analyze the JD for clarity, tone, and relevance, offering suggestions to enhance its appeal to potential candidates.
Novel Example:
Every YouTube user has a problem with link rot when videos become unavailable, disrupting access to valuable content and reducing the credibility of websites that rely on these links. Explore if AI can detect inactive YouTube links and automatically suggest or replace them with similar content, using metadata like titles, descriptions, and transcripts to ensure relevance. With similar content, saving you time and ensuring your research remains robust.
If there is value and interest, in future editions, I can cover functional use cases such as sales, marketing, HR, etc., though it seems there is a lot of published research by analysts and enterprise pilots (lots unfortunately are behind paywalls and NDAs).
2. Questioning/Asking
What makes a conversation flow smoothly, like in a Joe Rogan or Lex Fridman podcast, is the artful series of questions. Topical relevance, simplicity, coherence, and cohesion are the key properties that enable flows from conversations and probe deeper into the subject.
AI can mirror this process by engaging in dynamic, context-aware dialogues.
LLM Conversation Example 1:
Q: What are empirical judgments?
A: Empirical judgments are based on observation, experience, or experimentation.
Q: What are moral judgements?
A: Moral judgments are based on ethical principles and values.
Q: How do you classify this judgement – ‘History is written by winners’?
A: It is a subjective judgement.
Q: What makes it subjective as mostly we don’t have access to history written by losers?
A: Yes, you are right. It is not subjective.
Q: It is not empirical, moral or subjective then. Can we call it ambiguous judgement?
A: Yes and so on… (I had a great time on this)
LLM Conversation Example 2:
Q: What is Objectivism?
A: Objectivism is a school of thought that moral judgements are just as objective as empirical judgements.
Q: What is Relativism?
A: Moral judgements are not objectively true. True or false is relative to something else, say person or culture.
Q: A wrestler wins all rounds in a competition and enters into the finals. Before the final match, as she is found over-weight in the category and as per rules, the Objectivist view she should be debarred and disqualified from competition. Relativist view is she should get silver medal because she won all matches till the finals fairly. Only one of the two is possible: Which of these is morally right?
A: Different ethical frameworks can lead to different moral judgments in a given situation. Ethics is a complex field that encompasses various theories and perspectives on what is morally right or wrong.
3. Logical Reasoning and Chain-of-Thought
AI-Assisted Logical Reasoning for Online Shopping:
User: “I need a new laptop for gaming and video editing. My budget is $1500.”
AI:
In this example, the AI uses logical reasoning to understand the user’s needs, sets parameters (powerful hardware, dedicated graphics, budget), and finds suitable options.
AI-Assisted Chain of Thought for Meal Planning:
User: “I want to cook a healthy dinner for my vegetarian friend in 30 minutes.”
AI:
4. Making Judgments - to Discriminate, Identify, and Correct Unverifiable or False Statements
In a time of information abundance and attention scarcity, the ability to discriminate between fact and fiction, objective facts and subjective interpretations is crucial. AI contributes to the abundance and requires you to stay vigilant and guard against the following:
Factual Inaccuracies:
Fabricated Information:
Misinterpretations and Odd Responses:
To fact-check AI responses:
Becoming proficient in AI is much like learning a new language—it requires mastering a set of essential skills. By developing competencies in application thinking, questioning, logical reasoning, and making judgments, you can unlock the full potential of AI to enhance your work and personal life.
Maybe in this process, as posited by the Sapir-Whorf hypothesis, with AI, we will also become more capable.
These are extraordinary times to live—with abandon, and yet with caution so that the species of one and the species all become better.
I hope these insights help. Please subscribe, follow, and share.
Comments and questions are welcome.
Best wishes
II. Questioning / Asking
Good conversations flow from well‑sequenced questions—topical, simple, coherent, cohesive.
LLM Conversation Example 1
Q: What are empirical judgments?
A: Empirical judgments are based on observation, experience, or experimentation.
Q: What are moral judgments?
A: Moral judgments are based on ethical principles and values.
