2026 Curriculum

Artificial Intelligence — From Theory to Practice

A comprehensive AI knowledge overview for business leaders, executives and professionals. From theoretical foundations to real-world applications to a view of what's coming.

Start learning Contents

Program Contents

Curriculum contents

What AI Is & Key Terms

Definitions, core concepts and the vocabulary you need to master to understand artificial intelligence.

The History of AI

The journey from 1950 to 2026 — through the AI winters, the deep learning boom, and the generative AI era.

AI for Executives

Real-world use cases for CEOs: strategic decision-making, market forecasting, risk management.

AI Across the Business

Applications across 9 core departments — with concrete use cases and the time each one saves.

The Future of AI

AI trends in Vietnam and worldwide — agentic AI, AI-native, and the 2030 outlook.

Part 1 · Theory

What AI Is & the Core Terminology

Artificial intelligence isn't just a technology — it's a vast field of science with its own vocabulary. Getting the foundations right is the first step toward applying it effectively.

What is artificial intelligence?

Artificial Intelligence (AI) is the field of computer science that studies and builds systems capable of performing tasks that normally require human intelligence — including recognizing images, understanding natural language, making decisions, solving problems and learning from data.

AI is not a single algorithm but a collection of methods, techniques and mathematical models designed to simulate different aspects of intelligence. Under the most common taxonomy, AI is divided into three levels: Narrow AI — good at one specific task only, Artificial General Intelligence (AGI) — matching human ability across many domains, and Artificial Superintelligence (ASI) — far surpassing human ability in every respect.

As of 2026, every AI system in existence — including ChatGPT, Claude and Gemini — is Narrow AI, however sophisticated and versatile it has become.

The vocabulary you need to master

Machine Learning (ML)

A subfield of AI that lets computers learn from data without being explicitly programmed rule by rule. ML discovers patterns on its own from historical data.

Deep Learning (DL)

A subfield of ML that uses multi-layer artificial neural networks (deep neural networks) to process complex data such as images, audio and text.

Neural Network

An artificial neural network — a mathematical structure that mimics how biological neurons connect, transmit and process information in the human brain.

NLP — Language Processing

Natural Language Processing — the ability of AI to understand, analyze and generate natural language (English, Vietnamese, and so on) the way humans do.

LLM — Large Language Model

Large Language Model — AI models trained on enormous amounts of text data, able to write, analyze, translate and program. Examples: GPT-4, Claude, Gemini.

Generative AI

The branch of AI that can create new content (text, images, video, code) from prompts or input data.

Prompt Engineering

The craft of writing input instructions/questions for an AI to get the best possible results — an essential skill when working with LLMs.

Computer Vision

Machine vision — lets AI "see" and analyze images and video: recognizing faces, reading documents, inspecting product quality.

Agentic AI

Agent-based AI — systems that can plan, make decisions and take action on their own to accomplish a goal, rather than just responding to a single question.

RAG — Retrieval & Generation

Retrieval-Augmented Generation — a technique that combines searching an internal knowledge base with an LLM's text generation to produce more accurate answers.

Fine-tuning

The process of further training an existing AI model on an organization's specialized data to improve its accuracy.

Hallucination

An AI "hallucination" — when a model confidently states false or fabricated information with no supporting data behind it. An important limitation to keep in mind.

Part 1 · Theory

The Stages of AI's Development

From a lab idea in 1950 to a technology now reshaping the entire economy — a turbulent 75-year journey.

1950 – 1956 · Birth

Alan Turing and the Turing Test

Alan Turing posed the question "Can machines think?" in his famous 1950 paper. In 1956, the Dartmouth conference officially coined the term "Artificial Intelligence," opening up a new field of research.

1957 – 1974 · High expectations

The early era of optimism

The first AI programs appeared: solving math, playing chess, proving theorems. The perceptron (a single-layer neuron model) was invented. Researchers believed human-level AI was 20 years away.

1974 – 1980 · The first AI winter

Budget cuts, collapsing confidence

When the promises failed to materialize, AI research funding was slashed. Hardware and data limitations doomed many projects.

1980 – 1987 · Expert systems

Expert Systems boom in the enterprise

Expert systems were adopted widely across businesses. Japan invested heavily in its Fifth Generation computer project. AI returned with a commercial focus.

1987 – 1993 · The second AI winter

Expert systems hit their limits

Expert systems were too rigid, too expensive and hard to maintain. Once again, confidence and funding for AI dropped sharply.

1997 – 2011 · The renaissance

Deep Blue, the internet & big data

Deep Blue (IBM) defeated chess champion Kasparov (1997). The internet boom generated enormous volumes of data. More powerful GPUs paved the way for deep learning. IBM Watson won Jeopardy! (2011).

2012 – 2022 · The deep learning boom

AlexNet, AlphaGo & the Transformer

AlexNet (2012) revolutionized image recognition. AlphaGo (2016) beat the world Go champion. The Transformer architecture arrived (2017) — the foundation of ChatGPT, Claude and every modern LLM. GPT-3 (2020) proved the power of large language models.

2022 – 2024 · The generative AI era

ChatGPT, Claude & the AI race

ChatGPT launched in late 2022 and reached 100 million users in two months. A wave of other LLMs followed: Claude (Anthropic), Gemini (Google), Llama (Meta). Generative AI became a global phenomenon, and every industry began adopting it.

2025 – 2026 · The agentic AI era

AI that acts, not just answers

AI shifted from "answering questions" to "getting work done on its own." Agentic AI can plan, use tools and orchestrate multiple steps to reach a goal. 65% of businesses have already deployed agentic AI. Gartner forecasts that 40% of enterprise systems will integrate AI agents by the end of 2026.

Part 2 · What AI Has Already Done

AI for Executives & CEOs

AI isn't just a tool for the technical team — it's becoming an indispensable strategic assistant for CEOs and the C-suite. According to the Conference Board's 2026 survey, AI has moved from the periphery to the center of operating strategy.

88%

of businesses have
deployed AI (McKinsey 2025)

66%

of organizations saw
productivity gains from AI

1.7%

of revenue expected to be
invested in AI in 2026

$500B+

in global enterprise AI
investment in 2026 (Goldman Sachs)

Real-world use cases for CEOs

🎯 Data-driven strategic decisions

AI analyzes millions of data points from the market, customers, competitors and internal systems to build scenario models. A CEO can assess the impact of each decision before making it — from market expansion and M&A to pricing strategy changes.

Coca-Cola · Walmart · Pfizer

📊 Smart real-time dashboards

Instead of waiting for the weekly report, the CEO gets an AI dashboard that updates continuously — aggregating KPIs from every department, automatically spotting anomalies and flagging early warnings when a metric drifts off target.

General Electric · Accenture

🔮 Market & risk forecasting

AI uses predictive analytics to forecast market trends, customer behavior and supply-chain risk. The CEO shifts from reactive to proactive — seeing 3–6 months ahead instead of waiting for events to unfold.

Amazon · Goldman Sachs

🤖 Automating executive admin

An AI agent automatically takes meeting notes, builds to-do lists, tracks commitments, drafts follow-up emails and compiles reports. The administrative tasks that eat 30–40% of a CEO's time are now handled automatically by AI.

Financial Services · Air Carriers

👥 Talent & culture management

AI analyzes HR data to predict who is likely to leave, evaluate performance objectively, and identify skill gaps. The CEO gets a full picture of organizational health from data, not just gut feel.

HR Analytics · People Strategy

💰 Cost & ROI optimization

AI scans all operating costs, uncovers hidden waste and suggests smart cuts. It also evaluates the ROI of each project and investment, helping the CEO allocate resources more precisely.

IBM · Enterprise AI

💡 Key insight According to the BCG AI Radar 2026, CEOs fall into three groups: Followers, Pragmatists and Trailblazers. The Trailblazers — CEOs who treat AI as a strategic foundation rather than a technology project — achieve 2–3× the ROI of the other two groups.

Part 2 · What AI Has Already Done

AI Across Business Departments

AI is not IT's project alone. Below are real-world use cases already deployed and highly rated across 9 core departments in the enterprise, each with the specific time it saves.

🎯

1. Sales

Use Case	What it does	Savings
AI Lead Scoring	AI analyzes customer behavior, CRM data and purchase history to rank leads by conversion probability. Sales focuses on the highest-potential leads instead of calling at random.	~60% of lead-sourcing time
Sales Forecasting	ML analyzes historical data, seasonality and market trends to forecast revenue accurately. It cuts forecast error from 30–40% down to under 10%.	~5h/week on reporting
Drafting emails & sales proposals	AI automatically drafts personalized emails and builds commercial proposals based on each customer's specific needs. Sales just reviews and sends.	~70% of drafting time

📢

2. Marketing

Use Case	What it does	Savings
AI content production	AI generates blog posts, social captions, marketing emails and ad copy. Bain & Company reports businesses cutting content-production time by 30–50% with generative AI.	30–50% of content time
Segmentation & personalization	AI analyzes user behavior across your website, email and social channels to build hyper-personalized campaigns — the right person, the right message, the right moment.	+20–40% conversion
Campaign performance analysis	An AI dashboard automatically aggregates ROI across all channels (Google, Facebook, Email, SEO), flags underperforming channels and recommends reallocating budget.	~8h/week on analysis

👥

3. Human Resources (HR)

Use Case	What it does	Savings
Automated résumé screening	AI reads thousands of résumés, scores fit against the job description and ranks candidates. Integrity Staffing used an AI agent to interact with over 66,000 applicants, cutting response time from days to under 15 minutes.	~75% of hiring time
Automated onboarding	An AI chatbot guides new hires: answering questions about policies, procedures and company culture. IT/HR staff are meaningfully offloaded during that first month.	~60% of onboarding time
Predicting attrition	AI analyzes patterns — engagement level, overtime, feedback, tenure — to predict who is at risk of leaving in the next 3–6 months, letting HR intervene early.	-25% turnover

💰

4. Finance & Accounting

Use Case	What it does	Savings
Invoice automation & reconciliation	AI OCR reads invoices, extracts the data and reconciles it automatically against the ERP. IBM applied this internally, saving thousands of hours with processing 80% faster.	~80% of processing time
Fraud detection	ML analyzes millions of transactions in real time, detects anomalous patterns and flags suspicious transactions before they complete.	-50–60% fraud
Financial planning (FP&A)	AI automatically consolidates data from many sources, produces financial reports, forecasts cash flow and runs variance analysis — work that used to take 2–3 days now takes a few hours.	~70% of reporting time

⚙️

5. Operations

Use Case	What it does	Savings
Predictive Maintenance	IoT sensors + AI predict when machinery will fail before the breakdown happens. It cuts unplanned downtime by 35–50% in manufacturing.	-35–50% downtime
Supply-chain optimization	AI forecasts demand, detects supply-disruption risk and recommends optimal inventory levels. Amazon uses ML to optimize its entire same-day delivery logistics.	-20–30% inventory
Quality control with computer vision	AI cameras automatically inspect products on the production line, catching defects the human eye can't see, at inspection speeds 10× faster.	10× faster, 99% accurate

🎧

6. Customer Service

Use Case	What it does	Savings
AI Chatbot & Virtual Agent	An AI chatbot handles 60–80% of common questions 24/7 in natural language. Airlines use AI agents to rebook flights and reroute baggage automatically — freeing staff for the harder cases.	60–80% of tickets self-served
Customer sentiment analysis	AI analyzes voice, text and reviews to gauge satisfaction in real time. It catches unhappy customers before they leave.	-15–20% churn
Automated ticket triage & routing	AI reads the ticket, classifies it (technical/billing/complaint) and assigns it to the agent with the right expertise. It cuts manual triage time.	~90% classification accuracy

💻

7. Information Technology (IT)

Use Case	What it does	Savings
AI Code Assistant	AI coding companions (GitHub Copilot, Claude Code): autocompleting code, writing tests, reviewing code and catching bugs. Developers gain 30–55% in productivity.	30–55% of coding time
Incident detection & cybersecurity	AI monitors systems 24/7, detecting anomalies in network traffic, server logs and user behavior. It warns of cyberattacks before the damage is done.	60% of threats caught early
Automated IT help desk	An AI chatbot answers internal questions: "How do I connect to the VPN?", "Forgot my password?", "How do I install app X?". It cuts tickets to IT support by 40–60%.	-40–60% IT tickets

⚖️

8. Legal & Compliance

Use Case	What it does	Savings
Automated contract review	AI reads and analyzes contracts, spots risky clauses, compares against a standard template and flags what needs changing. Reviewing a 100-page contract drops from 2–3 days to a few hours.	~80% of review time
Regulatory compliance	AI tracks changes in laws and industry regulations, automatically assesses the company's level of compliance and warns when there's a risk of violation.	-50% violation risk
Legal document search	RAG-powered search enables semantic search across a huge legal document store. It answers "What's the indemnity clause in the contract with partner X?" in seconds.	~5–10h/week on lookups

🔬

9. Research & Development (R&D)

Use Case	What it does	Savings
Accelerating product R&D	AI simulates thousands of design variants, finding the optimal trade-off between cost, time and quality. Manufacturers use AI agents to optimize their new-product development process.	-30–40% cycle time
Research & patent analysis	AI reads and synthesizes thousands of scientific papers, industry reports and patents to identify technology trends and innovation opportunities. Pfizer uses AI to shorten drug R&D timelines.	~70% of literature-review time
AI-assisted experimentation	AI designs optimal experiments, predicts outcomes and reduces the number of physical trials needed. Especially effective in pharmaceuticals, materials and chemicals.	-40–60% experiments

⚠️ Important note The time savings shown are drawn from reports by McKinsey, Deloitte, Bain & Company and real enterprise case studies. Actual figures will vary with organization size, data maturity and quality of implementation. AI delivers the most value when integrated into existing processes, not when it fully replaces people.

Part 3 · What AI Will Do

The Future of AI — Vietnam & the World

We are at the early stage of a decades-long revolution. Below are the trends shaping the future of AI over the next 5–10 years.

🌏 The future of AI worldwide

🤖 Agentic AI — the era of autonomous AI

AI will shift from an "answering tool" to a "digital colleague" — planning on its own, orchestrating multiple systems, and completing work end-to-end. Gartner forecasts that 40% of enterprise systems will integrate AI agents by the end of 2026. By 2028, AI agents may manage entire complex workflows.

🧠 Multimodal AI — understanding every kind of data

AI will process text, images, video, audio and 3D data all at once. Applications: robots that understand the physical world, AI doctors that read both X-rays and medical records, virtual assistants that see and hear simultaneously.

🏭 AI-Native Enterprise

By 2028–2030, the successful company will be "AI-native" — every process, decision and product has AI built in from the start, rather than AI bolted onto legacy systems.

🔬 AI in science — accelerating discovery

AI has already helped uncover protein structures (AlphaFold), design new drugs and discover new materials. Over the next 5–10 years, AI will act as a co-scientist, accelerating research by 10–100×.

📐 AI regulation & ethics

The EU AI Act is now in force, and countries are building their own legal frameworks. The future demands AI that is transparent, explainable, fair and safe. Companies that build AI governance early will hold a competitive advantage.

🌐 AI Robotics — intelligence meets the physical world

AI robots will operate in factories, warehouses, hospitals and even homes. Combining computer vision + NLP + physical manipulation, AI robots become "colleagues" in the real world.

🇻🇳 The future of AI in Vietnam

🎯 The 2021–2030 national AI strategy

Vietnam aims to rank among the top 4 AI nations in ASEAN and the top 50 worldwide by 2030. Resolution 57-NQ/TW designates AI as a priority sector, with incentive policies for investment and AI-talent development.

🏢 A domestic AI ecosystem

Major corporations such as FPT, Viettel, VNPT and CMC are leading the build-out of Vietnam's AI ecosystem — from developing Vietnamese-language LLMs and AI-as-a-Service platforms to applying AI in digital government and smart cities.

👨‍💻 AI talent pool

Vietnam has major advantages: a young population strong in math and algorithms, and competitive labor costs. Vietnam is emerging as a leading AI-outsourcing hub in Southeast Asia, attracting investment from Silicon Valley and Singapore.

📜 AI law & legal framework

In 2026, Vietnam is drafting its own AI law and digital-technology regulations to balance innovation with responsibility. It's an important step toward a clear legal environment for AI businesses.

🌾 AI tailored to Vietnam

Vietnam-specific applications: smart agriculture (crop forecasting, pest detection), telemedicine for remote areas, personalized education, and smart-city management in Ho Chi Minh City and Hanoi.

🚀 Opportunities & challenges

Opportunity: leapfrogging technology, using AI to lift productivity across the whole economy. Challenges: fragmented data infrastructure, a shortage of senior AI experts, and the need to raise AI awareness at every level of management.

💡 The core message AI won't replace people — AI will replace the people who don't know how to use AI. The businesses and individuals who proactively learn, adapt and integrate AI into their work will hold a decisive competitive advantage over the coming decade. This isn't a technology race — it's a race of mindset.