AI Literacy: The Skill No One Taught You (But Should Have)
- Ricardo Brasil
- Jan 19
- 6 min read
By Ricardo Brasil | Lead Talks
You know how to use ChatGPT. Maybe you even play around with Midjourney on weekends. But are you AI literate ? The difference between using and understanding may be what separates those who will ride the wave of digital transformation from those who will be swallowed by it.
Functional Illiteracy in the AI Age
Here's an inconvenient truth: we're creating a generation of functionally illiterate AI users . People who use powerful tools without understanding their limitations, biases, or how they actually work. It's like driving a car without knowing there's a handbrake.
The Brazilian Artificial Intelligence Strategy (EBIA) identified this back in 2021. Among its nine pillars, one of the most critical – and ignored – is Pillar 4: Qualifications for a Digital Future . And no, it's not just about teaching Python to children (although that helps).
The Numbers That Frighten
EBIA presented data that should make you rethink that "I'll learn later" attitude:
📊 Only 15% of Brazilian graduates are in STEM fields (in China it's 40%) 📊 35 million Brazilian workers are at risk of automation by 2050
📊 500,000 tech jobs are expected to remain unfilled due to a lack of qualified candidates.
📊 12% of jobs in Brazil could be automated in the next 5 years.
But the worst statistic? 14% of AI professionals in Brazil are women . We are literally building the future with half the available talent.
What the Heck Is AI Literacy Anyway?
Forget the idea that you need to become a data scientist. AI literacy is about three layers:
1. Basic Literacy: Understanding the Basics
• Knowing the difference between AI, ML, and deep learning (no, they are not the same thing)
• Understand that ChatGPT doesn't "know" things, it predicts patterns.
• Recognizing when an AI is hallucinating... oops, confabulating
• Knowing the ethical and legal limits (hello, LGPD!)
2. Applied Literacy: Using it with Purpose
• Giving effective prompts (yes, this is a skill)
• Identify when AI helps versus when it hinders
• Understanding algorithmic bias and its consequences
• Knowing when a decision should NOT be automated
3. Critical Literacy: Questioning and Evaluating
• Detect deepfakes and disinformation
• Assessing whether an AI tool is ethical
• Understanding privacy implications
• Think critically about automation in your work.
EBIA Knew This (But Nobody Cared)
The document proposes something revolutionary: digital literacy at ALL levels of education . From early childhood education to high school. It's not about turning everyone into programmers, but into digitally conscious citizens .
The Brazilian National Common Curriculum Base (BNCC) already mentions "digital information technologies," but let's be honest: how many schools actually teach computational thinking? How many discuss ethics in AI? Algorithmic bias? Data privacy?
Plot twist: EBIA also talks about educating parents and guardians. Because it's no use for the child to learn at school if adults think "Alexa is magic."
The Three Levels of Digital Ignorance
Inspired by the EBIA classification, I identified three critical profiles:
🔴 Level 1: The Naive User
• Do you think AI is infallible?
• Shares data without thinking
• Does not question algorithmic recommendations.
"If Google said it, it's true."
🟡 Level 2: The Functional User
• Uses AI tools at work
• Knows how to make a decent prompt
But he doesn't understand what goes on behind the scenes.
"It works, I don't need to know how."
🟢 Level 3: The Literate User
• Understands limitations and biases
• Use critically
• You know when NOT to use AI.
• Questions automated decisions
The truth is: Most are at Level 1. Some reach Level 2. Very few reach Level 3.
The time is now!
In the Job Market
According to EBIA and IPEA studies, jobs that require creativity, empathy, and critical thinking are the most "AI-proof." But here's the trick: you need
Complement AI, not compete with it .
Professions at high risk:
• Telemarketing
• Data entry
Repetitive office work
• Basic financial analysis
Safest professions:
• Therapists and psychologists
Artists and creatives
• Teachers (the good ones, who facilitate learning)
Strategists who use AI as a tool
In Society
Facial recognition is already used in public security in Brazil. EBIA cites the case of the 2019 Micareta de Feira de Santana: 1.3 million faces captured . Of these identifications, less than 4% resulted in arrests.
But worse still: studies show that facial recognition systems have a much higher error rate for Black people. In the EBIA study of 151 prisons using facial recognition, 90.5% were Black people .
If you don't understand how it works, you can't question it. And if you don't question it, you passively accept a biased technology.
The Self-Literacy Roadmap
We're not going to wait for the government to implement EBIA in schools (spoiler: it's taking too long). Here's your action plan:
📚 Phase 1: Fundamentals (2-4 weeks)
• Take the "Elements of AI" course (free, in Portuguese)
· Read "Weapons of Math Destruction" by Cathy O'Neil
Watch "Coded Bias" on Netflix.
Understand the basics of machine learning (no programming required)
🛠️ Phase 2: Mindful Practice (1-2 months)
Use ChatGPT, but QUESTION the answers.
• Experiment with different prompts
• Test for biases: ask about professions and observe if there are stereotypes.
• Read the terms of use for the tools you use (yes, really).
🧠 Phase 3: Critical Thinking (continuous)
Follow AI ethics researchers on Twitter/LinkedIn
· Participate in discussions about regulation.
• Share knowledge (teach = learn more)
Question EVERYTHING that involves automated decision-making.
EBIA's Radical Proposal
The document proposes something few countries do: including concepts of linear algebra, calculus, and probability in high school as supplementary activities. Not for everyone to become a mathematician, but to understand the fundamentals of how AI makes decisions.
It also suggests:
✅ Technological training for teachers (urgent!)
✅ Digital literacy programs for ALL ages
✅ Interpersonal skills (soft skills) combined with technical skills
✅ Interaction between the private sector and education
The gap: This has been on paper since 2021. Where is the large-scale implementation?
Real-life cases that we should study.
While formal education is still lacking, the Brazilian public sector is already using AI in ways that impact your life:
🤖 Alice, Sofia, and Monica (TCU) - Analyze bids and audit contracts.
🤖 Victor (STF) - Classifies legal resources
🤖 Elis (TJ-PE) - Speeds up the processing of cases
🤖 PIÁ (Paraná) - Provides 380 government services via AI
Do you know how these AIs make decisions? What data do they use? Are they biased? You should know.
The Forgotten Generation
EBIA talks a lot about children and young people (right!), but there's a forgotten group: adults aged 30-50 . Those who are already in the workforce but will be working for another 20-30 years.
This group needs:
• Rapid requalification
Practical courses (less theory, more application)
• Reverse mentoring (young people teaching veterans)
Lifelong learning as a culture, not an exception.
OECD data: 6 out of 10 adults lack the skills needed for emerging jobs. In Brazil, this number is likely higher.
The Challenge of Diversity
Remember the 14% representation of women in AI? That's not only unfair, it's stupid from a strategic point of view .
Homogeneous teams create AIs with homogeneous biases.
EBIA recommends "diverse team composition" in terms of gender, race, and sexual orientation. Why?
1. AIs reflect their creators - Homogeneous team = biased AI
2. Different problems require different perspectives.
3. It's more profitable - Studies prove that diversity increases innovation.
The AI Literacy Test
Answer honestly:
• Do you know what a Large Language Model (LLM) is?
Can you explain algorithmic bias to your grandmother?
• Do you know when an AI is "confabulating" (lying without knowing it)?
Do you understand the difference between correlation and causation?
• Have you read the privacy policy of any AI tool you use?
Do you know what data your AI tools collect?
Can you tell when an image is generated by AI?
Do you understand the basics of how ChatGPT works?
Less than 4 ✓: Level 1 - Requires basic literacy
4-6 ✓: Level 2 - You use it, but you can delve deeper.
7-8 ✓: Level 3 - Congratulations, you are an exception.
The window is closing.
Here's the most important point: the window of opportunity is closing .
Companies are already requiring AI literacy as a prerequisite. Governments are regulating (slowly, but they are). Society is dividing between those who understand and those who will be impacted without understanding.
EBIA was visionary in prioritizing education. But the document alone changes nothing.
We need:
1. Individual action: Educate yourself NOW.
2. Institutional pressure: Demanding implementation in schools
3. Corporate responsibility: Companies training their teams.
4. Public debate: Discussing ethics, not just efficiency.
The Future Is Now (And You're Late)
I hate to be the bearer of bad news, but if you don't start today, it will be harder tomorrow. The AI adoption curve is exponential. The knowledge gap is too.
The good news? AI literacy doesn't require a new college degree or a 6-month bootcamp. It requires curiosity, critical thinking, and a few hours a week.
The bad news? Those who don't become literate will become increasingly dependent on systems they don't understand, making decisions they don't question, in a world they don't control.
The question isn't IF you'll need AI literacy. It's WHEN you'll start studying it.
Resources to get started TODAY:
· Elements of AI: elementsofai.com
• Coded Bias (Netflix)
• Complete EBIA: EBIA
Check out my other articles here:
Comments