Designing out hate
Social media used to be like the best Christmas party. Sadly not anymore. Lots of loud and angry have gatecrashed. What do we do about it?
Joel Bailey has been designing services since the early 2000s. He’s now Head of Product & Service at Arwen.ai — using AI to remove hate and make Social Media social again. In this article he shares some ways designers can contribute some seasonal goodwill by preventing the spread of toxicity and hate online.
It’s the season of goodwill, which is a such a good thing, as sadly, amongst other things 2021 will be remembered for, it has seen the online conversation get more toxic.
I’m old enough to remember when social media was all goodwill. A way to keep an ear to the ground, connect with others and a regular opportunity to laugh or have one’s interest piqued. Compare that to the following:
Microsoft have identified that each year, one-in five of the UK population are targeted with some form of online abuse. Two in every five of us have witnessed online toxicity.
A report by the Fawcett Society revealed that 66% of twitter users experienced abuse or harassment; 64% on Facebook, and 88% on Instagram.
These are big numbers! If you haven’t yourself been a victim of online abuse, you will likely know someone who has, or have witnessed someone getting a hard time online.
It’s rife, and we as designers have a duty to do something about it.
I’ve been a service designer since the early 2000s. Like many or you, I had many lucky years helping clients in a variety of sectors to build and enhance their own services. But since October I’ve turned my skills and experience to combatting online hate — joining Assimity to work on our Arwen product, which removes hate from social media.
In the last few months I’ve learnt that we all have the tools to reverse this toxic trend. In the spirit of Christmas I wanted to share, in the hope that you’ll each play your part.
Stopping people from hating is hard
I’ve been studying hate. It’s a human emotion, surging back and forth over the years. But as Nelson Mandela said: “No one is born hating another person because of the colour of his skin, or his background, or his religion. People must learn to hate, and if they can learn to hate, they can be taught to love, for love comes more naturally to the human heart than its opposite.” Mathew Williams has probably written the most coherently on how to counter it in The Science of Hate.
Designers can nudge people to reflect on the consequences of their action, prior to taking it. Erin Malone at the Anti Defamation League has created an invaluable set of design patterns to help you do this https://socialpatterns.adl.org/
Designers in the platforms can do more
The most common call amongst politicians and regulators is to get the platforms to act. We are seeing some progress here, with new time-out and key word filtering features being added by the networks. But compared to the scale of the problem, it feels a little tokenistic.
The networks are notoriously tight lipped about how their algorithms work to influence user behaviour, leaving plenty of room for people to speculate about what is preventing wholesale change. Exploring that would require a whole other article.
My Christmas request to designers working in the networks is to keep up the good work, find the solutions that the business model can accommodate. The emerging consensus is that negative content is more powerful at engaging eyeballs and driving profits than positive content.
Only a machine could like this state of affairs. But we’re humans. We’re human-centred designers. The machines should serve our designs
Protect the victims
This is the one we’re focusing on at Arwen. We believe everyone has the right to protect themselves from online abuse and take back control of their online voice. So we’ve designed and built a service that automatically removes toxic content and hate from social media conversations in under a second, around the clock.
It’s been a meaty design challenge — getting into the headspace of victims, classifying the 23 different types of hate that people receive, finding global providers who’ve created data models for our AI to run on and spot that hate. And then wrap that core product in an end-to-end service that helps explain an unpleasant problem and novel solution, and deliver it day after day, without rest.
My service design skills have been stretched. My product management skills have been stretched. My knowledge of artificial intelligence has grown ten-fold.
We’re not there yet. After all we’re making a new market. Few organisations or individuals have a budget for ‘toxic content prevention’, but this is a mission worth fighting for. The response from our customers has warmed my heart every week.
This Christmas, please think what you can do in your role to stem the growth of toxic content. Hate has a horrible tendency to creep into online spaces. But we all have the tools at our disposal to design hate out.
Happy Christmas
Joel Bailey is Head of Product & Service at Arwen.ai
