Unit 1: Introduction to Design thinking with AI tools

Design thinking is a human-centred, iterative problem-solving approach that pairs empathy with experimentation; when combined with AI tools it becomes a discipline for framing problems, generating options, and validating solutions at speed. This unit establishes the vocabulary (creativity, invention, innovation), the historical arc of the field, and the way AI project workflows now sit inside the design cycle.

• Human-centred: Solutions begin from the user's real needs, not the technology's capabilities.
• Iterative: Ideas cycle through build–test–learn loops rather than a single linear pass.
• Divergent then convergent: Widen the option space first, then narrow to a decision.
• Tolerant of ambiguity: Problems are reframed, not just answered, so early uncertainty is expected.
• Evidence-seeking: Assumptions are tested with prototypes and, in AI work, with data and metrics.

I. Understanding the AI Project Life Cycle

How an AI initiative moves from idea to deployment.

The AI project life cycle is the staged workflow that turns a problem into a monitored, deployed model.

• Problem scoping: Define the goal, success metric, and constraints — e.g. "reduce support-ticket routing errors below 5%."
• Data acquisition: Collect and label raw data; document sources and consent.
• Data preparation: Clean, de-duplicate, normalise, and split into train/validation/test sets.
• Modelling: Select an algorithm, train it, and tune hyperparameters on the validation set.
• Evaluation: Measure against held-out test data using metrics such as accuracy, precision, recall, F1.
• Deployment: Integrate the model into a product or API for live use.
• Monitoring & retraining: Track drift and performance in production; retrain when accuracy decays.

A. Link to Design Thinking

• Empathy ↔ scoping: Both start by understanding the user's real need before building.
• Prototype ↔ modelling: An early model is a testable artefact, exactly like a design prototype.
• Test ↔ evaluation: Metrics play the role that user feedback plays in design loops.

II. Introduction to Design Thinking Principles

The core commitments that govern every design decision.

Design thinking rests on a small set of principles that separate it from conventional planning.

• Empathy first: Observe users in context; document pains and gains before proposing fixes.
• Bias toward action: Build a rough version early rather than debate on paper.
• Reframing: Restate the problem to reveal a better one — "faster elevators" becomes "less boring waits."
• Radical collaboration: Cross-disciplinary teams surface options no single expert would.
• Fail early, fail cheap: Low-fidelity prototypes expose flaws before costs escalate.
• Show, don't tell: Communicate ideas through visuals and artefacts, not abstract description.

III. Understanding Creativity, Invention, Innovation

Three linked but distinct engines of new value.

These terms name successive stages: an idea, a working embodiment, and its adoption at scale.

A. Creativity

• Definition: The mental capacity to generate ideas that are both novel and useful.
• Anchor: Brainstorming 40 uses for a brick tests fluency and originality, not correctness.

B. Invention

• Definition: The first realisation of a creative idea as a working thing or method.
• Anchor: The transistor (1947, Bell Labs) was an invention — a device that had never existed.

C. Innovation

• Definition: An invention that reaches users and creates value in a market or system.
• Anchor: The transistor became innovation only when built into affordable radios and computers.

The three contrast sharply:

1. Creativity produces the idea (thinking).
2. Invention produces the artefact (making); innovation produces the impact (adoption).

IV. Kinds of Problems

Matching the method to the type of problem.

Design thinking targets problems that are ill-defined rather than merely complicated.

• Well-defined problems: Clear goal, known rules, single correct answer — e.g. solving 2x + 4 = 10.
• Ill-defined problems: Fuzzy goals, shifting constraints, many acceptable answers — e.g. "improve campus life."
• Wicked problems: No stopping rule, no true/false test, every attempt changes the problem — e.g. urban homelessness, climate policy.
• Tame problems: Bounded and repeatable; solvable with standard procedures.

Design thinking is most valuable for ill-defined and wicked problems, where reframing matters more than calculation.

V. Emergence & Evolution of Design Thinking

From craft practice to formal discipline.

Design thinking evolved as designers' tacit methods were studied and codified.

• 1960s: The "design methods" movement (Herbert Simon, The Sciences of the Artificial, 1969) framed design as a "science of the artificial."
• 1970s–80s: Robert McKim's Experiences in Visual Thinking and Stanford's product-design program stressed visual and hands-on ideation.
• 1991: IDEO is founded, popularising a repeatable, human-centred process.
• 2005: The Stanford d.school formalises teaching of the method to non-designers.
• 2000s onward: Tim Brown (IDEO) and Roger Martin extend it to business strategy and management.

The trajectory moves from individual craft → codified method → cross-domain management tool.

VI. Nature & Use of Design Thinking

What kind of thinking it is and where it is applied.

Design thinking is abductive, exploratory reasoning aimed at what "might be," not only what "is."

A. Nature

• Abductive logic: Reasons toward the best explanation or possibility, unlike deductive proof or inductive generalisation.
• Non-linear: Teams loop back to earlier stages as understanding deepens.
• Integrative: Balances desirability (users), feasibility (technology), and viability (business).

B. Use

• Product design: Shaping features around observed user behaviour.
• Services & healthcare: Redesigning patient journeys to cut friction.
• Education & policy: Prototyping curricula or public services with stakeholders.
• AI systems: Framing which problems are worth automating and how outputs reach users.

VII. Characteristics of Design Thinkers

The mindset traits that make the method work.

Design thinkers share a recognisable set of dispositions.

• Empathy: Genuine curiosity about users' lived experience.
• Optimism: Belief that a better solution exists and is findable.
• Experimentalism: Comfort building and discarding many prototypes.
• Comfort with ambiguity: Ability to work without a fixed brief.
• Collaboration: Preference for "T-shaped" skills — deep in one area, broad across many.
• Holistic sight: Seeing systems and interactions, not isolated parts.
• Iterative resilience: Treating failure as data rather than defeat.

VIII. Models of the Design Thinking Process

Competing frameworks for the same underlying loop.

Several models structure the process; they differ in labels, not in spirit.

A. Stanford d.school 5-Stage Model

• Empathise: Understand users through observation and interviews.
• Define: Frame a point-of-view problem statement.
• Ideate: Generate a wide range of solutions.
• Prototype: Build low-fidelity, testable versions.
• Test: Gather feedback and refine.

B. British Design Council Double Diamond

1. Diamond one (problem): Discover (diverge) then Define (converge).
2. Diamond two (solution): Develop (diverge) then Deliver (converge).

• Anchor: The two diamonds visualise alternating divergent and convergent thinking.

C. IDEO's Three Lenses

• Desirability: Do people want it?
• Feasibility: Can it be built?
• Viability: Can it sustain itself economically?
• Overlap: Innovation lives where all three intersect.

The models contrast as follows:

1. d.school emphasises sequential stages for teaching.
2. Double Diamond emphasises the rhythm of diverging and converging.

IX. Career Opportunities in Design Thinking

Roles where the mindset is a core requirement.

Design thinking underpins a growing family of professions across sectors.

• UX/UI Designer: Crafts interfaces grounded in user research.
• Design Researcher: Runs interviews, field studies, and usability tests.
• Product Manager: Owns problem framing and prioritisation.
• Service Designer: Maps and improves end-to-end service journeys.
• Innovation/Strategy Consultant: Applies the method to business transformation.
• AI Product Designer: Bridges user needs and model capabilities, deciding how AI outputs are presented and trusted.

Common thread: each role turns user insight into tested, deliverable solutions.

X. Monitoring AI Innovations

Keeping designed AI systems trustworthy after launch.

Monitoring is the ongoing surveillance of a deployed AI system and of the wider innovation landscape.

A. Monitoring the Deployed System

• Model drift: Input data shifts from training data, degrading accuracy; detect via rolling metrics.
• Performance metrics: Track precision, recall, latency, and error rate against live baselines.
• Bias & fairness audits: Check outputs across user groups for discriminatory patterns.
• Feedback loops: Route user corrections back into retraining data.

B. Monitoring the Innovation Landscape

• Trend scanning: Watch new models, tools, and regulations that reshape what is feasible.
• Ethical & regulatory watch: Track privacy law and AI-safety guidance affecting deployment.
• Competitive benchmarking: Compare capability against emerging alternatives.

Monitoring closes the design loop: production insight becomes the empathy data for the next iteration.