#1 - How to successfully design HCAI products?

Creative Currents Weekly Kick-off 🥳  

Welcome to our brand new product design newsletter! This is the place where we'll be sharing all the latest and greatest updates on designing awesome products that put people first.

For the first issue we want to focus on the research stream of Human-Centered AI (HCAI) as we believe that AI has the potential to revolutionize the way we live, work, and play, but only if it's designed with people in mind. That's why we're passionate about creating AI products that are not only innovative and cutting-edge, but also intuitive, user-friendly, and downright fun to use.

So, whether you're a seasoned AI product designer or just starting out on your journey, buckle up and get ready for a wild ride! We're excited to have you along for the journey.

Human-Centered Artificial Intelligence (HCAI)

 Here are a few reasons why human-centered AI is crucial:

👉 Ethical considerations: Human-centered AI emphasizes the importance of ethical frameworks, such as fairness, transparency, accountability, and privacy. It aims to mitigate biases, ensure AI systems make unbiased decisions, and provide clear explanations for their actions.

👉 User-centric design: Human-centered AI puts users' experiences and requirements at the forefront. It aims to create AI systems that are intuitive, user-friendly, and aligned with human values, preferences, and diverse cultural contexts.

👉 Collaboration and partnership: It recognizes the significance of collaboration between humans and AI systems. Human-centered AI seeks to build partnerships that leverage the strengths of both humans and AI, enabling more effective decision-making and problem-solving.

👉 Trust and acceptance: Human-centered AI strives to enhance trust in AI technologies. By involving users in the development process, addressing their concerns, and providing understandable and interpretable AI outcomes, it fosters trust and promotes widespread acceptance of AI.

👉 Social impact: Human-centered AI takes into account the broader societal impact of AI technologies. It considers factors such as job displacement, economic inequality, and the well-being of communities. By actively addressing these issues, it aims to ensure that AI benefits all members of society.

Picks from the Editorial Team 🤌

// 1 Explaining prices in Google Flights (LINK) - The case study discusses how the Google Flights team used the PAIR framework to identify pain points and create solutions that leverage AI to provide better user experiences. The team also used the framework to ensure that the AI-powered features of the platform were transparent, explainable, and trustworthy. Furthermore, the case study concludes with lessons learned from the Google Flights team's use of the PAIR framework, including the importance of user-centered design, clear communication of AI-powered features, and collaboration between designers, developers, and researchers.

// 2 The Human-Centered AI design process (LINK) - The author argues that AI technology should be designed with a focus on enhancing human experiences and meeting human needs, rather than solely focusing on technical capabilities. The article outlines a human-centered AI design process that consists of four phases: discovery, design, development, and deployment. In the discovery phase, the team identifies user needs and pain points. In the design phase, the team uses this information to ideate and create solutions that address these needs. In the development phase, the team creates and tests a prototype. In the deployment phase, the team launches the product and continues to collect user feedback to make improvements. Throughout all phases of the process, the author emphasizes the importance of collaboration, user testing, and transparency in AI design.

HOW to apply it in practice 🛠️ 

👉 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, talk to people, make use of methodologies like contextual inquiries to really understand the problem space (LINK)

👉 Augmentation vs Automation: Understand the context through talking to people. Tasks to automate are usually difficult or unpleasant. Augment tasks that people love and become more efficient through the power of AI (LINK)

👉 Define success: Define the UX and the role of the AI outcome, including the potential states (e.g., True negative, False positives etc.), for instance through a confusion matrix (LINK)

👉 Testing: Testing AI solutions is still one of the ongoing challenges. Due the complexity of AI systems and to get first results quickly, typical methodologies include Wizard of the OZ (LINK) or Paper prototyping (LINK)

HCAI research & key takeaways 📚

We strongly believe that the work of Verganti et al. (2020) “Design in the age of artificial intelligence” (LINK) provides great reflections regarding the unique value-add of applying artificial intelligence in new product development (NPD) and focus on the angle of “Design for AI”. The second research stream, “AI for design” covers all the elements around generative AI and how designers or creators apply a large-language model (LLM) in the product development process, which will be covered in future issues.

AI is not like any other digital technology. It does not just automate operations. It automates learning, which is the core of innovation. Thus it offers unprecedented opportunities to dramatically reduce the cost and time of developing a new solution.

👉 Scale and people centeredness: These machines have the capabilities to embed design rules that are inherently user-centered. In the case of Netflix, the application of supervised learning leverages a rich stream of data to each individual user. This focus can be scaled with no limitations on the number of users and the complexity of data, which results in personalised and a more effective user experience (UX).

👉 Scope and abductions: Products were designed for a specific industry, and with a specific target. For example AirBnB, which expanded their offering with travel experiences, by providing guests the possibility to take a horse ride on a beach or hire musicians. To enter this new industry, AirBnB leverages the same AI approach that powers the traditional hospitality service of AI.

👉 Learning and iterations: AI solutions are intrinsically iterative and deliver through loops. In the case of Netflix, each time a customer access the service, the firm activates a problem solving loop, which offers a new opportunity to further learn and provide an enhanced user experience (UX) over time.

Learn 🎓

• Introducing the HCAI Canvas (Canvas LINK)

• Lingua Franca - A design language for HCAI (Lingua LINK)

• Guidelines for Human Computer Interaction by Microsoft (Guidelines LINK)

• Community, Toolkits, Library, Tutorials (Toolkits LINK)

• Google PAIR Guidebook (Case studies, Workshop material) (PAIR LINK)

• Standford AI Index Report (2023 LINK)

Unclassifieds

• HCAI Research group - Ben Shneiderman (LINK)

• HAI - Stanford (LINK)

• IBM - HCAI (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