Listed as Free on Udemy when checked on 3 November 2026. No coupon or subscription trial is required for the listed lessons. Enrollment needs a Udemy account; availability can change.
About this course
A prediction about what people may do is different from a thoughtful decision about how to influence their choices. This course brings those questions together through behavioural science, cognitive biases and nudge theory. Its emphasis on ethical intervention makes it relevant to learners who want to examine the human assumptions behind AI-assisted education, marketing and decision systems, rather than treat a prediction as an instruction.
Yung-Hsiang Hu presents three sections and sixteen video lectures, with a listed duration of approximately one hour and fifty-two minutes and a June 2023 update. Each unit also includes a quiz. The description explicitly states that the course is conducted in English with Chinese subtitles, and the visible titles are bilingual. No prerequisite is specified, although familiarity with basic analytics can help connect the concepts with the intended applications.
| Official course | The Behavioral Sciences In the Age of AI (I) AI世代的人類行為干預 |
|---|---|
| Created by | Yung-Hsiang Hu |
| Platform | Udemy |
| Language | English |
| Level | Beginner — Beginner on Udemy |
| Listed duration | 112 minutes; 16 lectures across 3 sections |
| Certificate | Not included with a new free enrollment |
What you’ll learn
Explore how social influence, limited attention, defaults and biased beliefs can shape a decision. Compare automatic and reflective approaches to nudging, then consider transparency, consent and the interests of the people affected. Recognise that an intervention’s apparent effectiveness is a different question from whether its design and use are ethically acceptable.
Skills you’ll gain
- Bias recognition
- Choice architecture
- Default option review
- Nudge classification
- Transparency assessment
- Ethical question framing
Connect a prediction with its assumptions
The first unit introduces social proof, norms and the influence of other people. It then connects behavioural science with AI-driven predictions and choice architecture. Ask which behaviour is being discussed, what the proposed influence is and what evidence would support the explanation. A pattern in data can suggest a question, but it does not by itself establish the cause of a person’s choice.
An educational-robot lesson introduces an example involving learning alerts and experimental comparison. Treat that as a case to examine rather than a universal conclusion that robots equal or replace teachers. The population, task, comparison and outcome all matter when interpreting a study. This short course provides a discussion point; it does not establish that every educational AI intervention will work equally well in another setting.
Use bias concepts without labelling people
The second unit groups topics such as self-control, attention, loss aversion, default bias, social preference and biased beliefs. The source uses the phrase behavioural disorders, but these lessons should not be treated as clinical diagnoses or a way to label individual people. Their role here is to introduce possible influences on decisions and questions about how an intervention might respond to them.
Examples include entertainment recommendations, insurance decisions, automatic renewals and conformity to group expectations. These are contexts for analysing choice, not a recommendation to purchase a financial product or copy a particular social behaviour. Consider whether a default is visible, whether an alternative is understandable and whether a person can make an informed choice. The course’s examples do not determine what is appropriate for every individual.
Compare how an intervention works
The biased-beliefs lecture introduces anchoring, authority, confirmation and in-group biases. These terms can help identify alternative explanations and weaknesses in reasoning, including the analyst’s own. They are less useful when applied as shortcuts that dismiss anyone who disagrees. A more careful approach asks what information was considered, what was ignored and how a conclusion could be challenged by further evidence.
The final unit defines nudging and compares Type 1 and Type 2 approaches through automatic and reflective thinking. It also discusses cognitive load and transparency, then combines those distinctions in a two-by-two matrix. This gives learners a way to compare mechanisms. It should not be read as a rule that one intervention type always succeeds or that a classification alone settles whether an intervention is acceptable.
Make ethics part of the design question
Ethical questions run through the course rather than appearing only at the end. OECD guidance on applied behavioural insights likewise emphasises transparency, opportunities to object, privacy and appropriate consent. Ask whose interests a change serves, what participants understand and whether they can avoid an unwanted influence. An increase in a target metric does not by itself demonstrate benefit to the people whose behaviour is being measured.
For optional private practice, compare two fictional reminder messages about a routine learning task. Identify the default, the intended behaviour, a transparency question and one possible unintended effect. No personal data, real experiment, targeted marketing, account change or contact with learners is needed. This is a suggested reflection, separate from the course quizzes. Free lessons provide an introduction; they do not establish clinical expertise, research approval, advanced AI implementation skills or guaranteed behaviour change.
OECD tools and ethics for behavioural insights: Compare the course discussion with public-policy ethical considerations.
Related courses for your next step
- Add a related skillFree Client Conversations Course
Focus: Client discovery · Sales-call preparation
Beginner
- Add a related skillFree SQL Data Analysis Course
Focus: Database setup awareness · Workbench navigation
Beginner
Compare course focus, level and study time
| Course | Level | Focus | Study time |
|---|---|---|---|
| This course — Free Behavioural Science Course: Biases, Choice Architecture and Ethical Nudges | Beginner | Bias recognition · Choice architecture · Default option review | 112 minutes (Udemy listing) |
| Free Client Conversations Course: Prepare for Sales Calls with Integrity | Beginner | Client discovery · Sales-call preparation · Values clarification | 44 minutes (Udemy listing) |
| Free SQL Data Analysis Course: Start with Tables, Queries and Filters | Beginner | Database setup awareness · Workbench navigation · Table inspection | 117 minutes (Udemy listing) |
Frequently asked questions
Is the course taught in English?
The source explicitly describes English instruction with Chinese subtitles. Its visible lecture titles include both languages.
Are the bias topics clinical diagnoses?
They should not be used as diagnoses or labels for individuals. The course uses them to discuss possible influences on choices and intervention design.
Does an effective nudge automatically count as ethical?
No. Transparency, autonomy, privacy, consent and the interests of affected people need consideration alongside the measured outcome.
Questions & discussion
Share a useful question or correction. Comments appear after moderation. Please avoid personal or sensitive information.