Three things to improve acceptance of AI

Three things to improve acceptance of AI

U

BOLD INSIGHT

To truly deliver on the promise of AI, developers need to keep the end users in mind. By integrating three components of context, interaction, and trust, AI can be the runaway success that futurists predict it will be.

The bold future of UX: How new tech will shape the industry

Part 2  Three things to improve acceptance of AI

Artificial Intelligence (AI) is one of the hottest topics in tech right now. Conversations around AI inevitably lead to dreams of a world where a computer is predicting every need one might have and/or the impending doom of humanity through a SkyNet / Ultron / War Games-type scenario.

As entertaining as that discussion might be, instead I’m going to focus on what AI needs to do to become more functional and more accepted by society (that is, users). As it stands now, technology (including some of the advances in AI) seems to be advancing simply because developers want to see if they could build it. What my colleagues and I want to see, as user experience (UX) professionals, is meaningful advancements in AI that deliver functionality that is useful for users.

To meet this goal, I sat down with my colleague Gavin Lew, who has recently been talking a lot about AI, to identify three things that AI needs to be successful:

  • Context – At its core, AI is based on pattern-recognition. Once AI learns a pattern, it can make predictions about outcomes of similar patterns. However, while we’re giving AI the raw data it needs to recognize patterns, we’re not giving it the context in which to make good decisions. Our take on this is that we are doing a disservice to AI by not giving it the proper context.
    • An example of this is IBM Watson Health. IBM Watson for Oncology was fed data from the Sloan Kettering Cancer Center and then suggested treatments for various cancer types all over the world. It was able to suggest the correct treatment for lung cancer over 96% of the time in India. However, in South Korea, it was only correct 49% when suggesting treatments for gastric cancer. Why? Because South Korea’s treatments for gastric cancer aren’t in line with Sloan Kettering’s recommended treatments. In other words, Watson was lacking the context needed to suggest the right treatment approach.
  • Interaction – Our understanding of user interactions with AI is still developing. The user interactions of AI are largely still unknown to most. How is someone supposed to use AI? Is “use” even the right term when it comes to AI? Once it is fully realized, a complex AI system will entail the systems of a home, car, office, appliances, and personal tech gadgets, all talking to each other and exchanging information without the user having to actively do anything. Thus, the user is seemingly not doing anything to use AI, while the system itself is passing and parsing data behind the scenes.
    • Think ahead to the future where you have your own personal AI. Our interactions with AI may consist of nothing more than an offhand comment, essentially interacting with the AI without knowing that we’re doing so. For example, when I’m making breakfast and mutter to myself, “Almost out of milk,” a strong AI will know to remind me at an appropriate time to buy milk. Or maybe it will just take the initiative and order me a gallon of milk from the automated grocery service in my area and there will be a milk delivery timed for when I get home from work. Or maybe I don’t need to state that I’m out of milk for the AI to act…. perhaps finishing the gallon of milk is my passive interaction and the AI figures out what the next logical step is by ordering automatically.
  • Trust – Trust in AI has been a recent topic of discussion in the tech sphere. For people to want to use AI on a regular basis, they need to trust it. The early buggy interactions people had with Siri scared them away from voice assistants to the point that most have not attempted to try Microsoft’s Cortana. A new form factor (i.e., Alexa) finally encouraged people to give voice assistants (read: AI) a second chance, and it was more widely accepted and used.
    • But why? Because of trust. Trust is created when a question is asked, and the right answer is given, when a task is given and correctly performed, when a purchase is made and the correct product was bought, and, possibly most importantly, when personal info is kept safe.

Once AI has the three components of context, interaction, and trust, it will be much easier for it to hit the mainstream and be the runaway success that futurists predict it will be. Even if the above three pillars are never fully recognized, to truly deliver on the promise of AI to the end users, the developers of AI systems need to keep the end users in mind since the AI is ultimately being created to benefit them.

What are your thoughts on all of this? Comment below and let’s get a dialogue started!

This blog post part two of a series, The bold future of UX: How new tech will shape the industry, that will discuss future technologies and some of the issues and challenges that will face the user and the UX community. Read Part 1 that discussed Singularity and the associated challenges with UX design.

Singularity and the potential impact on UX design principles

Singularity and the potential impact on UX design principles

U

BOLD INSIGHT

If we are approaching a rapid technology shift as some experts predict, core UX design principles will have to be redefined to adapt to radically different interaction models.

The bold future of UX: How new tech will shape the industry

Part 1  Singularity and the potential impact on UX design principles

The times they are a-changin’. I know it’s a corny, overused refrain but I don’t think that it has ever been truer. Technology, as well as its impact on society, is advancing at a rapid pace, and that pace is only expected to accelerate.

Futurologist Ray Kurzweil believes that we are soon approaching a point where the computing power of tech exceeds the computing power of people. This “Singularity”, as it is called, will be fueled by a variety of emerging technologies, including artificial intelligence (AI), robotics, and nanotechnology, to name a few.

Once this Singularity hits, Kurzweil and other similarly-minded theorists believe that life will be unrecognizable to what we know today. He compares the difficulty of describing this post-Singularity society to someone today as being just as difficult as describing to a caveman how different life will be with bronze tools and agriculture.

Bringing us back to the present, how does this relate to UX?

My thoughts around this weird unknowable world of the future have started to stray toward design. Let’s think about AI and (by extension) robots. These two technologies have the potential to completely flip the paradigm of usability and user experience. The user should not have to learn how to use AI. AI is supposed to be the one learning: learning our habits and routines and learning what actions it should take in response to what’s happening around it. In UX research terminology, the user has become the stimuli and the stimuli has become the user. That is, the human is now the stimuli that the technology is learning to react and respond to.

But if you buy into the whole notion of Kurzweil’s Singularity, how do you design for a future that is (predicted to be) wildly different than anything we’ve ever known or could fathom? How can a UX designer implement traditional usability principles, such as effectiveness, efficiency, and satisfaction, or are these principles going to become a relic and left by the wayside as radically different interaction models emerge?

I’m going to tackle some of these questions in future posts in this series. Next topic: Artificial Intelligence!

What are your thoughts on all of this? Comment below and let’s get a dialogue started!

 

This blog post is the first of a series, The bold future of UX: How new tech will shape the industry, that will discuss future technologies and some of the issues and challenges that will face the user and the UX community.

Am I satisfied or stuck? The impact of ecosystems on household users

Am I satisfied or stuck? The impact of ecosystems on household users

U

BOLD INSIGHT

Manufacturers building an ecosystem of devices and services should design for both a separate, personalized experience and household or shared experience.

The idea of connected devices and a connected home fascinates me – I’m all for anything that makes my life more convenient! I have Alexa in pretty much every room of my house; she’s even in my car. However, as I expand my connected home network, I have struggled with setting up additional devices and services. Powering them on and account linking is generally simple; the hard part is getting everything to work together.

In the case of Amazon devices (e.g., Echo) and services (e.g., Music Unlimited), if you are single or start from one family/shared email address, the connected home ecosystem is pretty simple. You have one account tied to all devices, Prime, and streaming products and services. However, once you introduce one or more additional family members, things get much more complicated.

In my case, my husband and I each had our own Amazon account when we met. Even when we got married, it didn’t make sense to share an account because we liked being able to have personalized recommendations and to keep our purchase history separate. Some years later, I stumbled across Amazon Household that lets you tie separate accounts together so you can share Prime benefits. After linking our accounts, I thought we’d truly have a “household” account that would allow us to share all services and content. Unfortunately, you can’t share everything (i.e, purchased content (video) and certain subscriptions).

Fast forward to my first Echo devices – I was so excited to set them up and try them out! But when I tested out the List functionality (‘Alexa, add milk to the shopping list’), nothing showed up in my app. Why wasn’t this working?! After trying different things (and a little cursing) I realized that I had set up the devices with my husband’s Amazon account since were gifts for him and therefore I had to sign into the Alexa app through his account, not mine. With Amazon Household, I didn’t think it would matter which account the Echos were tied to, but it does.

What is technically easy to set up, actually requires a high cognitive load each time I set up a device or access content because I have to remember which account I used for what. I currently have:

  • Amazon Prime account with my email address which is linked to my husband’s Amazon account (with his email address) so he can get Prime
  • Alexa app on my phone but signed in using my husband’s Amazon account for Echo devices and lists
  • Amazon Music Unlimited account signed in using my husband’s email address
  • Roav VIVA Alexa-enabled device in my car that requires me to sign into my Amazon app with my husband’s email address to get access to Music Unlimited, but to shop and see my recommendations, I must sign back into the Amazon app with my email

One could argue that I should have been more intentional when setting up all these devices and services. But in the moment I was so excited to get these things working that which account to use was the last thing on my mind. I’ve questioned if I should suck it up and start all over with a family account. But what would I gain? Possibly an easier setup process going forward and one account for everything, but lots of effort up front to reset everything. And what would I lose? Personalized recommendations, purchase privacy, and time!

Netflix and Hulu have overcome this multi-account hurdle with their ‘profile’ platform which generates separate watch lists and recommendations. Admittedly, they are much simpler systems with limited components.

There are huge benefits to having an ecosystem of devices and services in a home, whether it’s Amazon, Google, Apple, etc. The consumer benefits by (generally) having a seamless experience of integrating the devices and services and working from a similar interface or set of commands used across multiple devices. For the manufacturer, the benefits of having its ecosystem in a home means more loyal customers since, for the consumer, it can be difficult or impractical to try new devices when the home is entrenched in one ecosystem.

Many connected device manufacturers have created a great set-up-and-use experience with plug and play devices and simple mobile apps. However, manufacturers should think beyond the experience of a single user. Consider how a couple or family would set up, purchase, use, and add to the ecosystem. Consider couples who come with individual personal accounts and those who create a family account together. Also consider early adopters who have tied accounts to early versions of the system – ensure there is support to improve their experience as devices or new features are added. Some questions to ask include:

  • What content would users want to keep separate: purchase history, recommendations, watch/wish list, etc.
  • What content would users expect to share: purchased content, services, etc.
  • Can established individual accounts be tied together to form a true “household” account?

Ultimately, as the foothold of any ecosystem gets stronger, the user can either feel satisfied and happy or stuck and frustrated. And that feeling (satisfied or stuck) becomes associated with the brand.

Confirmation Fatigue: Enough with the reminders already!

Confirmation Fatigue: Enough with the reminders already!

U

BOLD INSIGHT

While confirmations and reminders are essential and usually well intentioned, they are annoying. Organizations must uncover if and how their customers will use reminders and confirmations during user testing. Confirms should add to the user experience, not leave users feeling frustrated and tethered.
I just switched phones from iOS to Android. In re-creating my new phone environment, I was irritated by how many apps and settings I had to confirm. Later in the day, I got an email from my dentist office asking me to confirm my appointment. Then I made a reservation for dinner on an app, yet the restaurant still called (and emailed!) to confirm. What’s with all this confirmation?

If you believe the hype, digital media was supposed to make my life easier and more convenient. It hasn’t really. It’s just making me tired and crabby. It’s what I call ‘Confirmation Fatigue’. Confirmation Fatigue, along with its close cousins ‘Alert Fatigue’ and ‘Acceptance Fatigue’, are making me a slave to my devices: I get paranoid that if I fail to confirm my appointment it might be canceled (I am told, though I’m not sure as each hold an unclear consequence).

These are but a few examples in a normal day:

  • If my bank suspects that some credit card transaction I’m making is fraudulent, it will deny the charge and send me a text message to confirm (and I usually don’t see the message until after the embarrassment and I’ve moved on to another vendor).
  • Just about every site I visit in the European Union (EU) makes me agree to (i.e., confirm) that it’s OK to drop a Cookie on my machine as the EU Cookie Law requires me to consent. It’s as tiresome as it is annoying.
  • My pharmacy not only sends me texts, but also emails and phones to make sure I pick up my prescriptions. They have a mechanism to allow you to opt out of these aggressive reminders, but it has not worked for me.
  • My doctor knows that if I don’t confirm, I am less likely to show. So, they cyberstalk me to remind me and harass me until I confirm.
  • After requesting a rideshare, I still find that half of the drivers call me to confirm. Why?
  • Every time I start my car, my navi app makes me confirm that I won’t use my touch screen while driving. Every. Time. There are even web sites devoted to overriding the navi certain cars. (Yes, we’ve met your lawyer, we won’t hold you responsible, just let us use the app!)
  • Apple is notorious for confirming software updates to aggravating frequency.
  • Lastly, while writing this piece, Chrome asked me to confirm updates. Sigh.

How did we get here?

We got here, in many ways, because the user interface was made too easy. The apps allow us to quickly and conveniently set appointments because the machines are so good at reminding us, we are relieved of the burdening of remembering (I.e., we forget). While these reminders and confirmations are at one level reassuring, it tethers us to the tech in ways I wish we weren’t. I am the digital bitch – just the weakest link in the supply chain. And they know it.

A generation ago we might have to look up the phone number in the phone book, dial, speak to someone during business hours, and arrive at an agreed upon time. We had to make effort and that effort committed us to what we were asking: doctor’s appointment, plumber visit, or lawncare. Since the effort to agree on appointment dates and times are getting increasingly frictionless, our commitment and our memories are more fleeting. So the machines take over and set up alerts and reminders requiring us to (re!)confirm that commitment.

What can be done to reduce the need for reminding and confirmation?

There is no easy answer. Confirmations are here to stay and largely good, but they are annoying. The real question is: Does everything need confirming…and confirming…and confirming?

I have started to ignore requests in situations where I feel I am not required to respond, like my dentist or the pharmacy. In other situations where I am compelled to respond (like the bank example), I must give in to Skynet and confirm.

To bubble up critical confirmations from the noise, one approach might be to make people think and commit more deeply. For instance, and take a random example, if you are asked to confirm whether it’s OK to send a message to every mobile phone in the State of Hawaii about a missile launch, perhaps you should have to put in some effort. Be forced to think about things: for instance, “In order to confirm sending the SMS message, you have to enter the square root of the atomic number of Manganese”.

Obviously, I’m kidding, but the reality is, technology only works if it works for people. A little research can uncover if and how the user wants to be reminded – how strongly, how often, and where. So rather than inciting feelings of annoyance, it promotes a mental release of the burden of forgetting something in our cluttered minds or alerts us before we make a mistake which totally changes the experience.

Pin It on Pinterest