#3 - Design Thinking & AI [When two dinosaurs meet and create value]

👋 Welcome to the third issue of creative currents! Get ready to be inspired and informed as we dive into the latest trends, tips, and insights from the world of Design Thinking & AI. Whether you're a seasoned product designer or just starting out, this newsletter is your go-to resource for all things around creating human-centered products which people ❤️

So, sit back, relax, and let's embark on this creative journey together! 

Design Thinking & AI - why should you care?

👉 In the process of designing AI applications - human needs are often not considered: For instance in a real-time coaching application, sales agents rather prefer a f2f coaching and don’t want to share their secret patterns to success. Or within the medical domain, doctors do not trust the AI application although the provided output reveals a high level of accuracy.

👉 When it comes to AI, the majority of organizations face similar challenges:

1. Unrealistic expectations from leadership about accuracy or what kind of tasks AI is able to take over

2. No data collection strategy which ensures data quality, availability of data or the collection of relevant data

3. No data pipelines are installed, which collect, label, version and continuously update the data

4. No measurable business objectives introduced

5. Disconnect between IT, data science and engineering

Resources of AI/Data Science challenges:

Why 87% of AI/ML Projects Never Make It Into Production—And How to Fix It - (d2iq LINK)

Why 90 percent of all machine learning models never make it into production -(Towards Data Science - LINK)

Why only one out of 10 data science projects get to production - (Mesh AI LINK)

How can Design Thinking add value?

👉 Problem framing: Design Thinking emphasizes understanding and defining the problem accurately from the user's perspective. This is crucial for AI development because identifying the right problem to solve is essential.

👉 User-centric focus: Design Thinking places the user at the center of the design process, emphasizing empathy and understanding their needs, desires, and pain points. Incorporating Design Thinking into AI development ensures that AI systems are designed with a deep understanding of the end-users.

👉 Ethical considerations: AI raises important ethical considerations around issues such as bias, privacy, and transparency. Design Thinking encourages interdisciplinary collaboration and stakeholder engagement, enabling diverse perspectives and expertise to address these ethical concerns.

👉 Iterative prototyping and testing: Design Thinking's iterative approach aligns well with the development of AI systems. AI algorithms often require training data and feedback loops to improve their performance. Design Thinking promotes prototyping and user testing at various stages of development, allowing AI teams to gather user feedback, refine algorithms, and iterate on the solution.

👉 Creative ideation and innovation: AI development can benefit from the creative problem-solving techniques of Design Thinking. Design Thinking encourages brainstorming, ideation, and exploration of alternative solutions. Applying Design Thinking to AI opens up possibilities for innovative uses of AI technology and enables the discovery of novel approaches to solving complex problems.

Picks from the Editorial Team 🤌

// 1 The symbiotic relationship between design thinking and AI (Medium - LINK) -  AI is widely used for tasks like disease diagnosis, fraud detection, and quality control. HCD, on the other hand, is an approach to problem-solving that puts people at the center. It involves understanding user needs, ideation, prototyping, and testing to create user-centric solutions.

AI and HCD can work together in several ways. AI technologies can be developed to be more user-friendly, ethically responsible, and improve existing HCD processes. HCD can inform the usability and usefulness of AI systems and help improve the development of AI technologies. AI can also benefit from design thinking principles by considering the user, iterating and testing, being open to feedback, simplifying designs, questioning assumptions, and considering the context and usability.

// 2 Applying Design Thinking to Artificial Intelligence. Why Should You Use It in Your AI-Based Projects? (LINK Nexocode)

One of the challenges for design thinking in artificial intelligence is, that there is no universal approach for it, but the Design thinking methodology proves that it can be used for effective and useful approaches to artificial intelligence products or services development. It’s definitely not an easy task, and certain organizations might have problems with it.

HOW to apply it in practice 🛠️ 

👉 Baseline

Find the intersection of user need and AI strength - At the beginning of the design process, understand early which problem you are trying to solve. Maybe it’s not even an AI problem. Map existing experiences (Experience Mapping LINK), talk to people (Interviews LINK), understand the processes (Contextual enquires LINK, Service Design Blueprint LINK) within the problem space.

From an organizational perspective, to understand the level of AI readiness, carry out an AI assessment (LINK):

Results of the AI assessment are highlighted in the following matrix. Each quarter represents an archetype, defined through the achieved outcomes and their experience of deploying AI applications.

👉 Intelligent element

Analyse, which intelligent behavior should be supported. Understand if AI can solve the problem in a unique way and when AI is probably better and when not. Define how success looks like, for instance through the application of the confusion matrix (LINK) and the following worksheets (LINK).

👉 Strategic options

There are different ways to define the overall AI and product strategy. One approach could be through the Hax's Delta Model (LINK), which is a customer-centric strategy framework that places the customer at the center of management decisions. It emphasizes building strong customer relationships and aligns strategy and execution through adaptive processes. The Triangle diagram represents three distinct approaches for implementing Delta Model strategies:

• System Lock-in: Aim for network dominance, acquiring complementors' share, driven by system economics

• Best Product: Focus on consumer happiness through cost reduction and product differentiation

• Total Customer Solutions: Offer personalized portfolios of products and services for individualized customers

👉 Technology

In parallel of the use case activation, enabling the organization plays a major role in the development of the PoC, which includes the assessment of the AI and IT/Data infrastructure capabilities. To activate use cases within organizations, the following phases provide a rough indication of execution:

• Pressure test use case (Go - or No Go decision, 1-2 weeks)

• Build analytics PoC and confirm the use case (8 - 12 weeks)

• Test and iterate MVP and prepare for scale (3-6 months)

• Scale up and continuously improve (6+ month)

For simple prototyping, ML5.JS (Friendly Machine Learning For The Web - LINK) provides ready trained models to run in the browser.

👉 UX of AI

Defining the user experience of AI application cannot be better explained than by the Google Pair guidebook, which provides detailed explanations and principles in the format of AIM FOR and AVOID (LINK), for instance “Define Errors and Failure”:

Further chapters explore - User Needs + Defining Success, Data collection & Evaluation, Mental models, Explainability + Trust, Feedback + Control, Errors + Graceful Failure (LINK)

Learn 🎓

• Design Thinking for AI (LINK)

• AI X Design (LINK)

• The Art of AI (LINK)

• AI fundamentals by IBM (LINK) - The AI Essentials Framework is a specific grouping of activities to work through to align your team on strategy for an AI experience

AI design sprints have emerged as a response to the growing need for efficient and effective AI development processes. These sprints combine the principles of design thinking and agile methodologies to tackle the unique challenges posed by AI projects:

• Nexocode (LINK)

• 33a (LINK)

• Siili (LINK)

Struggling with a design challenge? Let’s jump on a call

Share the Creative Currents Newsletter with at least five friends in your network and Book 👋 ([email protected]) your Free Consultation Session with a design expert 

Support the Creative Currents newsletter

• Forward it to your product design friends and recommend them to Subscribe

• Have product design related topics like events, newsletters, tools, jobs, articles or anything you'd like to share with our subscribers? Share Now

• Sponsor the next edition and become part of the network