SUTD Design and Artificial Intelligence Latest News 2026 | Supercharged DAI, Human-Centred AI and 2027 Admissions

SUTD Design and Artificial Intelligence (DAI) latest news in 2026 starts with a major curriculum change: from September 2026, the Bachelor of Science in Design and Artificial Intelligence uses its Supercharged DAI structure, introducing two tracks — Societal and Systems Intelligence and Human-Centred Intelligence. The School’s 2026 news also highlights research about when AI can observe children’s behaviour without making clinical judgements, and an experimental ACPAgent designed to explore the ethical limits of AI acting on a person’s advance care preferences. Students choosing a 2027 degree should distinguish the specialised DAI bachelor’s from SUTD’s separate Computer Science and Design degree and its new Human-centred Domain Intelligence second major.

The university’s live graduate programme catalogue confirms a 31 May 2027 deadline for the September 2027 Master of Science in Technology and Design (Human-Centred Design), while the undergraduate application pages checked on 12 November 2026 still displayed the 2026 cycle. Do not simply change the year on an older undergraduate deadline. This independent eduKateSingapore guide gives quick explanations, real 2026 evidence, curriculum comparisons, application guidance, fees and student exercises designed to help readers make informed decisions.

Quick answers for SUTD DAI in 2026

January 2026: all Freshmore students begin studying Design·AI

The 6 January university release announced a revised common first-year curriculum called Freshmore.

From the AY2026 cohort onward, incoming SUTD students encounter a specially designed sequence, Innovating with Design and AI, across the first three terms. The university describes using AI in stages of design such as understanding users, gathering context, developing concepts, prototyping and evaluating alternatives.

This change applies to the university’s shared first-year foundation, including learners who will later major in Architecture and Sustainable Design, Computer Science and Design, Engineering Product Development or Engineering Systems and Design.

That is different from enrolling in the specialised Bachelor of Science in Design and Artificial Intelligence, which develops deeper human-centred and social systems learning after Freshmore.

Why connect AI with the entire design process?

A student may be able to generate a polished image or paragraph in a few seconds without identifying a problem that someone genuinely needs solved.

Design begins earlier than the picture: Who is the user? What are the constraints? What evidence is available? What happens when the proposed solution fails?

AI can help explore variations, but the design team must decide which variations are appropriate and how to test them.

For example, a tool could generate many versions of a clinic waiting-room interface. A design decision still depends on how actual visitors read instructions, whether different languages are supported and whether the interface is accessible to people with disabilities.

The technological tool can accelerate the work, but cannot eliminate the need to investigate the real situation.

What is human-centred design?

Human-centred design begins by understanding the people affected by a product, service or system.

It may involve interviews, observation, usability studies, prototyping and iterative testing. A good designer also considers people who are likely to be overlooked by an average-user assumption.

A service used by an older adult with visual difficulties and a student with strong digital skills may need different controls or forms of feedback.

Human-centred does not mean ignoring technical constraints. It means treating human needs as essential technical requirements rather than optional decoration.

A useful school-level project

Imagine a class designing a homework organiser for students who have different school schedules.

The first idea might be an attractive weekly calendar. Students should then ask whether learners can update unexpected changes, whether reminders are excessive and whether the tool makes private information visible to other people.

A useful project documents what was tested and which assumptions proved wrong.

Students can use AI to generate draft layouts if their teacher permits it, but the finished work should accurately disclose tool use and demonstrate independent reasoning.

February: SUTD supercharges the dedicated DAI degree

The 29 January curriculum announcement and 12 February university news report explained that the DAI bachelor’s was being broadened beyond conventional technical design.

The revamped programme welcomes applicants with strong interests in Economics, Business, Finance, Psychology, Philosophy and human–AI interaction, while continuing to expect academic readiness in Mathematics and the Sciences.

The programme combines foundational Computing, AI and design with social-science methods and applied projects.

That matters because modern technical decisions involve institutions, incentives and people. An accurate predictive model may still cause harm if it is deployed without fair access or a clear explanation of who is accountable.

Is DAI the same as Computer Science and Design?

No. The dedicated DAI degree is designed around the intersection of Design, AI and human or societal understanding.

Computer Science and Design (CSD), associated with SUTD’s ISTD pillar, concentrates more deeply on computing foundations, algorithms, software and specialised technical systems.

Both may study AI, but their compulsory and elective structures differ. A student planning to be a software engineer should inspect CSD’s mathematics and systems depth, while one interested in social-impact AI services may find DAI’s approach relevant.

The most accurate comparison looks at actual modules, capstones and technical skills rather than which programme name sounds more fashionable.

The two new DAI tracks

The official Supercharged DAI track catalogue presents two options, with track-specific core subjects and at least seven approved elective courses under the stated pathway rules.

Students must follow actual prerequisites and any requirement to develop depth in a sub-specialisation. A published elective list is not permission to enrol in every advanced course immediately.

Societal and Systems Intelligence (SSI)

SSI is structured around Economics, Business and AI.

Its listed core includes Principles of Economics in the Age of AI, with electives addressing Microeconomics, Macroeconomics, Econometrics and Data Analysis, Causal Inference for Policy, Game Theory and System Design, International Trade, digital platforms, Finance and organisational processes.

This track is attractive for students interested in how technological decisions change markets, cities or public services.

For example, a digital financial service may be technically fast but inaccessible to people without suitable devices. Its design therefore depends on business incentives and public equity as well as code.

Why Economics matters in AI design

Economics examines choices under constraints and how incentives affect behaviour.

A business may deploy AI to reduce customer-service costs. That could be economically sensible in some respects, but the company must also consider customers who cannot use the automated service, the cost of errors and whether staff are available when escalation is needed.

A design proposal that counts only employee hours saved is incomplete if it ignores customers’ time or service failures.

The student should learn to identify both measurable outcomes and relevant social consequences.

A systems exercise: congestion at a clinic

Suppose a clinic wants to reduce waiting times. A simple app could tell all patients to arrive early, but that might make waiting rooms more crowded.

A better analysis looks at appointment capacity, consultation duration, no-show patterns, accessibility and staff schedules.

An AI forecast could help, yet it must be tested against actual operations and appropriate privacy rules.

The system is not successful merely because average wait time decreases if certain patients now face much longer delays.

This illustrates the SSI approach: combine computational, economic and human factors.

Human-Centred Intelligence (HI)

HI focuses on Psychology, Philosophy and Human–AI Interaction.

The published track list names Mind, Machines & Design as the track core.

Electives include Introduction to Psychology, User Interaction Design Fundamentals, Technology and the Self, Persuasion and Decision-Making, Collective Behaviour, Health Communication, Computational Models of Cognition, Moral Philosophy, AI Ethics and Equitable Technology.

A student interested in how users understand AI systems may find these subjects especially relevant.

Why Psychology is not just user preference surveys

Psychology can investigate attention, memory, decision-making, motivation and behaviour through empirical methods.

A survey may tell a designer that people like a new interface. It does not automatically prove the interface improves learning or reduces errors.

Researchers may need controlled comparisons, objective task measurements and follow-up testing.

A student should be prepared for quantitative analysis and careful evidence, not merely discussions about personalities.

Why Philosophy is relevant to AI

A model can produce a prediction, but a system’s designers still need to decide whether using that prediction is justified.

For example, an AI tool might rank applicants for a job. A technically accurate model can still raise fairness, privacy or accountability questions.

Philosophical reasoning helps clarify what counts as fairness, which rights are at stake and why an organisation should explain its decisions.

These are not problems solved automatically by a higher prediction score.

A design exercise: an AI tutor that always agrees

Imagine a learning chatbot that praises every student answer.

Students might feel supported, but they could become more confident in incorrect reasoning.

A human-centred redesign would consider how to give precise correction without humiliating the learner, and how to prompt independent explanation.

The right educational outcome is not simply user satisfaction in the moment; it is improved understanding over time.

14 July: AI observes parent–child behaviour, but cannot replace experts

The SUTD research release dated 14 July 2026 describes work led by Assistant Professor Kenny Choo and colleagues.

The research examined whether multimodal AI could identify behavioural cues associated with joint attention during parent–child interaction.

Joint attention occurs when a child and another person share interest in an object or activity while recognising that the other person is engaged.

The researchers worked with three experienced speech-language pathologists to understand how experts observe gaze, actions and vocalisations.

The AI was trained to identify and describe such cues before trying to interpret their quality.

What did the system achieve?

The system reached approximately 81% accuracy in identifying selected observable cues under the study’s evaluation.

However, performance declined substantially when it attempted to make expert-level judgements about the *quality* of joint attention.

The researchers also found that experienced clinicians sometimes agreed about what they observed but differed in how they interpreted its significance.

This is a crucial distinction: a reliable description is not the same as a reliable clinical interpretation.

Why the result matters to child development

Developmental assessments involve context, background, individual variation and specialised training.

A gaze direction or gesture may have different meanings depending on the child, activity and interaction.

An AI timeline could eventually help a clinician review relevant moments, but it should not diagnose a child or automatically recommend treatment without appropriate validation and professional oversight.

The researchers explicitly called for larger, more diverse testing before considering broader use.

What parents should and should not conclude

The paper does not establish that parents can upload home videos to any general chatbot and receive a valid diagnosis.

Families should not share identifiable child recordings with an unauthorised service or treat an AI label as a clinical finding.

If there are genuine concerns about communication development, seek appropriate professional advice using authorised healthcare channels.

The educational value is in understanding why expert judgement and careful data protection remain essential.

A simple school-research lesson

Suppose students observe how often classmates raise their hands during a lesson.

Counting raised hands is an observation. Concluding that a child lacks interest because they raised their hand less often is an interpretation.

The student may be shy, may already understand, may need extra time or may have other reasons.

A good researcher separates the count from the explanation and investigates uncertainty before making a claim.

This is the same logic that protects child development research from premature conclusions.

14 July: ACPAgent asks whether AI can express someone’s care wishes

The SUTD technical feature on ACPAgent describes an exploratory research prototype presented at the 2026 CHI Conference on Human Factors in Computing Systems.

Advance care planning allows people to record their preferences for future medical decisions. The research team investigated whether an AI agent could help someone articulate those preferences, especially when a person might later be unable to speak.

Across four workshops, 15 participants considered scenarios about treatment and care, training or challenging the agent as they reflected on their values.

The study is an ethical and design investigation, not a hospital-deployed system legally authorised to make end-of-life decisions.

Why 86.7% agreement is not proof of safety

The team reported agreement between people and the AI agent in approximately 86.7% of the decisions in the study.

But researchers cautioned that a high agreement score might reflect the system learning preferences, a participant deferring to persuasive language or an AI model simply being agreeable.

For sensitive personal decisions, agreement is not the same as informed consent, authorship or genuine autonomy.

The study also found cases where participants changed their initial choices after hearing the agent’s argument.

This raises a deeper question: if a system provides the words through which someone expresses a value, can it unintentionally influence the value itself?

How did participants respond?

People disagreed with the agent over family responsibilities, affordability and uncertain recovery prospects.

The model sometimes failed to capture emotional or contextual concerns that mattered to the participant.

The team explored roles ranging from a low-autonomy tool helping someone express preferences to a more independent proxy.

The researchers considered a middle position — a tool that helps communicate a person’s preferences without taking over the human decision.

The actual law, healthcare practice and responsibilities of clinicians remain separate from the experimental design exercise.

A high-stakes design lesson

An AI tool should not claim to know what is best for a person merely because it predicts a likely response.

The user needs transparency about stored information, the ability to challenge a recommendation and appropriate control over what is shared.

A future clinical application would require legal, ethical, data-protection and practical validation.

A student considering DAI should find this a powerful example of why human-centred design cannot be reduced to making an interface convenient.

July: Professor Yow Wei Quin’s recognition

The SUTD DAI and HASS news pages reported on 16 July 2026 that Professor Yow Wei Quin was elected a Fellow of the Gerontological Society of America.

Her research interests include language, multilingualism, human development and ageing-related questions.

Academic fellowship is a recognition of professional contribution, not an automatic endorsement of every technological product or a personal scholarship for all DAI entrants.

The broader relevance is that an AI-centred degree can benefit from scholars who understand how people communicate, age and learn.

Technology affects people differently across ages and social environments. A well-rounded team should investigate those differences rather than presume one generic user.

The Human-centred Domain Intelligence second major

SUTD’s 2026 HDI second-major page explains a change in name and design.

The previous Design and Artificial Intelligence Second Major is now described as Human-centred Domain Intelligence (HDI).

It is designed for eligible students across SUTD’s five undergraduate majors: Architecture and Sustainable Design, Computer Science and Design, Design and Artificial Intelligence, Engineering Product Development and Engineering Systems and Design.

The goal is to combine domain expertise with AI knowledge and human-centred design.

What do non-DAI students study through HDI?

The HDI page describes a route for other majors involving design studios, AI courses and HASS Human-Centred Design subjects.

For example, a student studying Engineering Systems may gain more explicit training in user research and AI’s social implications.

However, the actual curriculum comprises prescribed combinations, and students must satisfy prerequisites and programme conditions.

A second major is not automatically a second complete bachelor’s degree.

What about DAI students pursuing HDI?

DAI students can use the second major to develop additional domain expertise, such as Healthcare or Manufacturing, within the approved structure.

This makes sense because a graduate who knows human-centred design but also understands a particular industry can work more effectively with specialists.

For example, hospital software design requires familiarity with clinical workflow and professional constraints beyond general UI knowledge.

The university determines which domain designations and course combinations are approved.

Should everyone take a second major?

No. An additional major may be useful when it supports a clear goal, but it consumes time and academic capacity.

A student could instead focus on a strong primary-degree project, an internship or deeper advanced technical electives.

A longer list of credentials does not automatically produce stronger capability or employment outcomes.

Plan with an academic adviser rather than simply collect the most impressive titles.

How the DAI undergraduate curriculum actually progresses

The Supercharged DAI curriculum page lays out an eight-term learning structure.

Students begin with the university-wide Freshmore year:

This is not a purely Arts degree without quantitative content. Mathematics, Computing and methods remain central.

Why Calculus and Linear Algebra matter

AI models often represent information numerically and learn through optimisation.

Calculus helps explain changes, gradients and how optimisation behaves. Linear Algebra helps describe vectors, matrices and transformations used in data and models.

A student who finds these subjects difficult can develop the necessary foundation through appropriate study, but should not expect AI tools to replace mathematical understanding.

A model can return a plausible answer even when its assumptions are wrong.

Why data collection matters before modelling

A team might want to use AI to predict which residents need extra help during a heatwave.

If its data include mainly younger smartphone users, the resulting model could overlook older people without comparable digital records.

The issue begins with the data and research design, not with the model’s sophistication.

A human-centred curriculum should teach students to ask who is represented and who is missing.

What is the graduate route in Human-Centred Design?

SUTD’s Master of Science in Technology and Design (Human-Centred Design) is a separate postgraduate programme connected to HASS’s educational work.

The current programme is 12 months full-time or typically 24 months part-time for eligible students, with a possible extension under specified university conditions.

Its academic structure comprises eight courses and 96 credits, including design core and specialist subjects.

It is appropriate to consider for some existing degree holders or professionals seeking human-centred research and design skills. It is not a direct substitute for a first degree.

The confirmed September 2027 deadline

The live SUTD graduate master’s listing states that applications for September 2027 close 31 May 2027.

Programme-specific admission requirements still apply. The MTD Human-Centred Design application page describes the prior undergraduate qualification and possible English-proficiency documentation.

The postgraduate fee published for AY2026/27 was S$49,050 for Singapore citizens and PRs in the relevant full-time category, and S$54,500 for international applicants, inclusive of GST. The correct 2027 amount should be checked before accepting an offer.

What jobs does the master’s prepare students to explore?

SUTD lists fields including user research, UX design, product management, design consulting and consumer insights.

Employment depends on previous preparation, projects, technical skills and market conditions. A master’s is not a guaranteed promotion or professional licence.

An applicant should examine the actual course and project work and determine whether it addresses a clear capability gap.

When can an undergraduate applicant apply for AY2027?

The SUTD undergraduate admissions page checked on 12 November 2026 still displayed the 2026 intake.

The qualification-specific application timeline must be consulted for updated 2027 dates.

A past 19 March 2026 A-Level deadline, or a January–March 2026 period for other qualifications, cannot be presented as a verified AY2027 deadline by changing its year.

Applicants should use their actual A-Level, polytechnic, IB, NUS High School or other recognised qualification route when the new cycle is published.

An undergraduate application checklist

  1. Confirm the named qualification you will present.
  2. Read the Supercharged DAI curriculum, including Mathematics and Computing.
  3. Compare SSI and HI and decide which questions interest you.
  4. Understand how SUTD’s Freshmore major selection works.
  5. Prepare official examination and diploma records.
  6. Document genuine design or research work clearly, identifying your own contribution.
  7. Monitor the official 2027 admission window and any interview or supplementary-material instructions.
  8. Review tuition, housing, scholarships and possible overseas activities.
  9. Submit through the authorised SUTD system and save confirmation.

An applicant should not pay an intermediary that claims to guarantee selection.

Tuition and financial aid: use the correct cohort

The SUTD undergraduate fee schedule lists S$13,500 per academic year for eligible subsidised Singapore citizens entering in AY2026.

Under SUTD’s cohort policy, that annual amount is fixed for the standard duration of the cohort’s undergraduate degree.

For a four-year illustration, 4 × S$13,500 = S$54,000 in tuition, excluding miscellaneous charges and living costs.

This is not an automatic AY2027 price. The newly entering cohort needs to check its own official fee table.

What costs should families include?

Budget for mandatory student charges, meals, local transport, housing if needed, computing devices and approved project materials.

A student may also consider overseas learning or work attachment requirements under programme rules.

Scholarships and bursaries are subject to eligibility. Loans are financing arrangements and may need to be repaid.

University staff recognition or a research award should not be mistaken for a personal student scholarship.

For broader comparisons, see eduKate’s university financial-aid guide.

What can DAI graduates realistically do?

A graduate may investigate roles in human-centred AI product development, UX research, technology consulting, design strategy, responsible AI governance, digital services, business innovation or further specialist study.

The right role depends on their track, technical skills, project evidence and actual hiring requirements.

Someone interested in clinical tools may need to work with licensed healthcare professionals; a DAI degree does not grant clinical professional status.

Someone interested in finance may need to comply with financial-services regulation; a university elective in Investing does not itself authorise regulated advice.

Students should learn to distinguish academic knowledge from occupational permissions.

Full Subject-Based Banding and the longer route to DAI

Singapore’s Full Subject-Based Banding concerns G1, G2 and G3 secondary-school subject levels. These are not direct SUTD undergraduate admission categories.

A learner later presents an accepted post-secondary qualification and must satisfy the university’s academic criteria.

English and Mathematics support communication and quantitative work. Science, Computing, Humanities and Art may contribute different useful perspectives to design.

Parents should not treat a PSLE score as a permanent judgement about future university access.

A child can develop human-centred habits early by listening, observing, asking precise questions and checking whether a proposed solution helps real people.

For the earlier educational pathway, see eduKate’s Full SBB guide.

Three reader scenarios

A JC applicant who enjoys Psychology and AI: compare the Human-Centred Intelligence track with a general Psychology programme and with Computer Science and Design. Investigate whether you prefer researching people, building technical models or joining both through design.

A polytechnic graduate interested in AI business systems: examine SSI’s Economics and digital-platform subjects, but prepare for Calculus, programming and data methods.

A mid-career professional thinking about UX: the September 2027 Human-Centred Design master’s has a confirmed 31 May 2027 application deadline. Check your prior degree, workload and fees separately from first-degree DAI admission.

Frequently asked questions

What changed in the SUTD DAI degree for AY2026?

The university introduced the Supercharged DAI curriculum with stronger Social Sciences and Humanities integration and two new tracks.

Which are the two Supercharged DAI tracks?

Societal and Systems Intelligence, focused on Economics, Business and AI; and Human-Centred Intelligence, focused on Psychology, Philosophy and human–AI interaction.

Does DAI mean learning AI without Mathematics?

No. Freshmore and subsequent study include programming, Calculus, Linear Algebra, probability, statistics and AI methods.

Is DAI the same as Computer Science and Design?

No. They share some technologies but emphasise different foundations and academic objectives.

What happened in the July child-interaction AI study?

The system identified selected behavioural cues with approximately 81% accuracy in the reported tests, but clinical interpretation remained a human expert responsibility.

Did ACPAgent become a legally authorised medical proxy?

No. The 2026 study used a research experience prototype and 15 participants. The team identified major legal, ethical and technical barriers to deployment.

Does 86.7% agreement prove an AI is trustworthy?

No. It was one result in an exploratory study and can reflect deference or an agreeable model as well as preference alignment.

What is the HDI second major?

A revised Human-centred Domain Intelligence second major, formerly called the DAI second major, with approved pathways across all five undergraduate majors.

When do Human-Centred Design master’s applications close for 2027?

The graduate catalogue lists 31 May 2027 for the September 2027 intake.

What is the 2027 undergraduate DAI deadline?

The SUTD undergraduate pages reviewed still displayed the 2026 exercise. Use the official updated qualification-specific calendar instead of guessing.

Can Full SBB students eventually apply for DAI?

Potentially through accepted later post-secondary qualifications and the university’s current entry criteria. Secondary G1/G2/G3 levels alone do not determine the result.

Official sources and connected eduKate reading

Editorial review, 12 November 2026: A curriculum announcement is distinguished from a separate specialist degree, 2026 research prototypes are not represented as legally approved clinical systems and the 2027 graduate deadline is not confused with still-unverified undergraduate dates. All findings are qualified by study conditions. No manufactured search volumes, admissions guarantees or career promises are made.

Explore the connected learning guides

Choose the question that brought you here. Open one useful guide, try a small task, and stop when you have what you need.

Take one question further

The same learning habit can travel across subjects, while each subject keeps its own methods. These routes help you notice a difficulty, understand one part of it, and return to something you can do.

A word is familiar, but using it is difficult.

Move from recognising a word to retrieving it in a new context. Understand vocabulary plateaus.

Try it without the guide: Choose one word you already know. Close the guide and use it in a new sentence. Explain why it fits; try another context tomorrow.

A piece of writing has ideas, but the reader loses the thread.

Make the order of events and the links between sentences clear. Explore composition writing.

Try it without the guide: Choose one short paragraph. Read the relevant explanation, close it, and revise the paragraph. Ask someone to tell you what happened and why.

The Mathematics seems familiar, but marks still disappear.

Find the first point where the working stops being reliable. Find Secondary 4 A-Math mark leakage.

Try it without the guide: For a Secondary 4 A-Math question you have attempted, locate the first uncertain line. Repair that step, then try a comparable question without the worked answer.

A Science fact is remembered, but the explanation is incomplete.

Connect the evidence to a scientific idea and the resulting change. Follow the Primary Science learning route.

Try it without the guide: Choose a familiar Primary Science example. Explain the evidence, the idea and the result without notes. Then change one condition and explain your prediction.

Two accounts of the world seem to disagree.

Check the question, source, date and evidence before combining claims. Explore the World Knowledge research library.

Try it without the guide: Take one claim. Find the source best placed to support it, note its date, and state what remains uncertain. Return to your original question.

There is plenty of help, but independence is hard to see.

Check what the learner can understand and do after support is removed. Understand how education works.

Try it without the guide: Choose one small task the child has practised. Agree on a calm, brief attempt without prompts. Use what happens to choose one next step, then stop.

For the structure behind these connections, read the eduKateSingapore runtime manifest and the eduKate ecosystem boot contract. The reader map describes public navigation; those manifests preserve the wider ownership and return rules.