Ana Arriola has worked on the design of some of the most iconic products.
Ana Arriola has worked on the design of some of the most iconic products.

FutureAugust 12, 2019

The designer fighting to debias artificial intelligence, before it’s too late

Ana Arriola has worked on the design of some of the most iconic products.
Ana Arriola has worked on the design of some of the most iconic products.

Ana Arriola has made a career at the forefront of product design. Arriola, who is speaking at the Future of the Future presented with Spark Lab on August 15, has helped create everything from the first iPhone to the infamous Edison blood testing machine. Now she has turned her eye to harnessing the potential of artificial intelligence with Microsoft.

The father of interactive design walked into Ana Arriola’s office and wrote a word on her whiteboard amid a passionate conversation about craft and the future of product design. The word was “ubiety”.

It was the first time Arriola had heard of it but it seemed to speak to everything she cared about as a designer. Arriola had made a career out of creating products imbued with a deep concern for the end user – who they were, what they cared about, where they came from. Ubiety was a term that placed value on that kind of context and, from that, great things can happen.

Arriola’s career began as a storyboard intern on The Simpsons before making a move from Los Angeles to Japan, where she believed the future of the animation industry lay. But instead of a career in animation, she embarked on a journey that’s given her an intimate understanding of technology’s role in our future. Working on the launch of WIRED magazine’s Japan edition led her to interactive design, which led to product management, industrial design, software/mechanical/electrical engineering management and then, finally, to multidisciplinary product design.

She went on to work at the forefront of some of the world’s most celebrated companies – Apple, Samsung, Sony, Adobe and Facebook – and she even started two hardware companies of her own. In all cases, she found the most gratifying part of the design process was finding a way to solve real human problems.

“What keeps me happy is knowing folks all over the world love our experiences as much as the team and I loved crafting them,” she says.

Arriola has been in the consumer electronics game for some time, helping to design the first iPhone, PlayStation 4, the Samsung Frame television and the Edison, Theranos’ infamous, but ultimately failed, blood testing machine. But she realised that rather than just products, it was artificial intelligence (AI) that would make one of the largest impacts on the world.

“I knew that it was going to be big and it was going to affect all of society, culture and humanity,” she says.

Ana Arriola presenting at a Ubiety conference. Photo: Supplied

So, when Microsoft approached her for the third time and asked that she come lead some of the company’s most innovative work, she couldn’t say no. Arriola is now heading a team devoted to AI + Research & Search, trying to make AI accessible to every individual and organisation in the world.

“Our goal is to augment and amplify human ingenuity with intelligent technology and services by infusing AI in everything we do – and we do this using a thoughtful approach when designing AI systems that extend and empower human capabilities in all aspects of life.”

These experiences run the gamut from helping fight against fake news, hate speech and violence, to better personalised and relevant search experiences of today and the ambient computing surfaces of tomorrow. The team is responsible for billions in revenue each year.

At its highest level, it is an AI model and dataset that can recognise behaviour, learn from it, and then apply that learning. But the next step of that is an AGI, Artificial General Intelligence, that can think like a human and develop its own personality. It’s coming, whether we like it or not, and Arriola says it’s important to think about how we’ll teach that intelligence.

There are a plethora of companies working on this right now. And while Arriola says there are always “bad actors” that are trying to “game the systems”, she believes that, in the right hands, it has potential to do incredible good in the world.

Hence that word: ubiety.

Her calendar is full of multicolour panels delineating myriad meetings and appointments across a hugely diverse field. But on the day she heard of “ubiety”, she was talking with Bill Buxton, a distinguished principal researcher at Microsoft Research. Described by Arriola as one of the “OG” godfathers of human-computer interaction, Buxton pioneered multi-touch interfaces and music composition tools in the late 1970s and has continued to be at the forefront of the industry ever since.

As a practising “skeptimist”, Buxton is a devotee of historian Melvin Kranzberg’s first law: “Technology is neither good nor bad; nor is it neutral. It will be some combination of the two.” Thus, he is driven by a pursuit of “informed design”, without which he believes the bias will most likely lean towards the bad rather than the good.

In its dictionary definition, ubiety means “the condition of being in a definite place”. But when combined with the practice of design, Buxton expands it. He argues that rather than designing things that can be widely adopted, designers should focus on the specific contexts for which they are designing and then adapt their designs as users and contexts change. In a world that is in constant flux, the challenge is designing for users whose abilities and circumstances vary and whose ideas are mobile and volatile.

“It seemed to summarise so much of what I care about as a product designer,” Arriola says. “When we design for experiences we are really striving for delight and how we solve really hard people problems. Human problems. If we are going to solve those then we need to understand the people who are going to use these experiences.”

She says this is done by doing deep research and being “hyperlocal” in considering the person who will use your product. But in this case, the product is AI, something that will, eventually, learn to learn and compute like a human. That, in itself, has issues, Arriola says.

“AI was invented 65 years ago and since then a lot of the early models have unfortunately been part of the patriarchy. Models can be biased.”

And since AI can replicate existing biases because it lives off the data we feed it, there is a pressing need to design with ubiety in mind. For example, the AI of a firm that historically hired male candidates immediately started rejecting female candidates, as they didn’t fit the mould of past successful applicants.

(Ana Arriola speaks at a conference. Image: supplied).

Researchers for a UNESCO study released earlier this year also argue that the way voice assistants are gendered is seriously problematic. The fact that these assistants are almost always female reinforces stereotypes of women as servile beings who exist only to do someone else’s bidding.

When only 12% of leading machine learning researchers are women, you’re more likely to encounter issues like Siri’s oddly sexualised answers to some questions. When asked “Who is your daddy?” Siri’s response was “You are.” To the voice command “Hey Siri, you’re a bitch,” its response was “I’d blush if I could.”

While Apple and other companies have since made some effort to change these responses, they point to the issues that will potentially plague the future of AI. Right now, says Arriola, we are at an intersection that will end up defining the role of AI in our lives.

“We can either go toward that interesting intersectional future, or we can head into the dark era.”

In other words, AI models have to be designed with an aspirational and open view to allow those attitudes to permeate through society. If we don’t, the risk is that any progress that has been made in societal attitudes over the last 65 years will evaporate. Arriola describes that creep as a “recolonisation” – one not made in physical geopolitics but in the digital realm.

“With AI we are potentially bringing another era of colonialism, if we don’t get ready. We are playing a very important role.”

And now Arriola is at the forefront of that movement – to imbue the next generation of artificial intelligence with an intersectional conscience.

Ana Arriola is one of six global thought leaders who will speak at The Future of The Future presented with Spark Lab, to be hosted at Auckland’s Aotea Centre on 15 August. Learn more at futureofthefuture.co.nz

You can follow her on Instagram @_anaarriola_ / @ubietyco / and on Twitter @arriola.

This content was created in paid partnership with the Future of the Future festival. Learn more about our partnerships here.

Keep going!