2019 is another pivotal moment for design.
While it seems design's heyday has been happening for the past several years, industry practitioners and leaders are still recognizing its immense need to play a pivotal role in shaping our economy, technology, and society.
This article is a reflection of IBM’s Cloud Data and AI Design Strategy team’s perspective on the trends emerging from SXSW 2019 based on the following talks. These aren’t immediately new trends, but their continued messaging from across industries highlights their growing importance.
1. Ethics and AI
2. Actual Design for Real AI
3. The Age of Smart Information
4. Bias In, Bias Out: Building Better AI
Design sits “between the intersection of society and technology,” creating a new challenge for designers as technology increasingly evolves. Designers now have to be “more like an orchestrator of data and systems” than the previous palette-pleasers of the past. For the sake of this article, we define a designer as:
Designer (n): anyone who gives form to new ideas or influences the end-user experience.
Major technological trends and new organizational initiatives have been driving design to play an evolving role in the products, systems, and services being made, but now designers are facing a new challenge: designing for ethics, AI, and mixed reality (XR).
Challenge #1: Designing for Ethics and AI
Design sits at the intersection between technology and society, and as more and more software applications incorporate some kind of artificial intelligence, the makers of these programs will have to consider the broader implications of their work and potential unintended consequences.
Companies like IBM and Google are publishing new principles for AI ethics and Stanford’s opening a new Institute for Human-Centered AI Ethics to establish guidelines around designing and building autonomous algorithms.
Response: For better teams and artificial intelligence, we need better inclusion of diversity.
Talks on both AI and Innovation had the same message: biased data-in leads to biased data-out, and homogeneous teams under-perform those that are more diverse. This means that in order to get the kind of unbiased AI we want, and to create better team collaboration, we need to address diversity and inclusion.
Secondly, software teams need to communicate and hold themselves accountable to new guidelines for the responsible use of AI.
Responsible AI includes:
- Accountability
AI designers and developers are responsible for considering AI design, development, decision processes, and outcomes. - Value Alignment
AI should be designed with consideration of the norms and values of your user group. - Explainability
AI should be designed for humans to easily perceive, detect, and understand its decision process. - User Data Rights: AI should be designed to protect user data and preserve the user’s power over access and uses.
- Fairness: AI should be designed to minimize bias and promote inclusive representation.
Challenge #2: Designing for Mixed Reality (XR)
Voice and augmented reality (AR) are already here. They’re streamlining human-computer interactions and slowly putting keyboards and desktop computers to rest. We’ve evolved over millions of years to interact with our world three-dimensionally and communicate vocally. Now software and hardware are now catching up.
With integrated cloud platforms from companies IBM, Microsoft, Google, and Amazon offering voice services a-la carte, the barrier to adopting and implementing such robust technology has dropped dramatically for smaller businesses of all kinds, requiring more teams to re-imagine what it means to communicate information architecture and human-computer interactions in a dimensional space with voice commands.
This presents new states for anyone designing a system to consider:
• is the computer listening?
• is it processing my request?
• did it just learn something new?
• is it uncertain?
• does it need my permission?
Response: Invest in exploring new mediums for big data visualization and interaction.
For larger companies looking to differentiate by enabling customers to gain deeper insights from big data, new visualization and human-computer interaction techniques present a real competitive advantage opportunity.
Companies like IBM are modernizing what data visualization looks like to help teams better work alongside machine learning algorithms and big data to identify insights, trends, and patterns in a sea of information. Yesterday’s graphs for information sense-making will look like still photographs next to tomorrow’s motion pictures.
This is because information is rarely stagnant. Whether its web traffic, revenue, weather patterns, or political policy, the data evolves over time. Exploring all this new data through flat two dimensional screens leaves out the reality of a 4-dimensional world (3 physical dimensions and time).
In Summary
The times they are a changing. We already knew that. But the unexpected elements of artificial intelligence are giving rise for new ethical considerations not previously needed when machines couldn’t autonomously learn for themselves. As machine learning algorithms evolve, designers need to imbue their work with foundational ethics to mitigate implicit bias, feedback loops, prejudices, and unintended negative consequences.
As our world becomes increasingly digitized as information, those who can capture and make the most sense of it will have a competitive advantage. Mixed reality interactions will enable teams to analyze data in 4-dimensions, bringing clarity to trends, patterns, and insights.
