Instagram, Nextdoor, and “Be Nice” Nudges

One of the first pieces of empathy-building tech* I wrote about was an algorithm built to recognize when comments on a newspaper story went off the rails. It was a tough story to place because it was hard to understand and even harder to explain. (I’m forever grateful for good editors!) The gist was that a group of researchers wanted to see if they could cultivate an environment in the comment section of a controversial story that would facilitate good, productive conversation. Their work eventually turned into Faciloscope, a tool aimed at detecting trolling behaviors and mediating them.

Like many research projects, it’s kind of hard to tell what happened after the initial buzz – grants change, people move, tech evolves, etc. All’s been pretty quiet on the automated comment section management front for a while, but over the past few months that’s begun to change. Now we can see similar technology popping up in the apps we use every day.

randalyn-hill-Zl2yVDTGByY-unsplash.jpg
Photo by Randalyn Hill on Unsplash

Earlier this year, Head of Instagram Adam Mosseri announced that the app would soon have new features to help prevent bullying. The official plan was released yesterday, and it boils down to one new function: Restrict. According to Instagram, “Restrict is designed to empower you to quietly protect your account while still keeping an eye on a bully.” It works letting you approve Restricted people’s comments on your posts before they appear – and you can decide to delete or ignore them without even reading them too, if you want. You won’t get notifications for these comments, so it’s unclear to me how you’d know they happened unless you went looking for them, which hopefully you aren’t doing, but let’s be honest… we all do that

Anyway, what about direct messages? DMs from Restricted people will turn into “message requests,” like what already happens when someone you don’t know sends you a message. The sender won’t be able to see if you’ve read their message.

Inexplicably, Instagram also used this announcement to tell us about its new “Create Don’t Hate” sticker, as if that’s an anti-bullying feature… when it’s literally just a sticker you can put on your story. So… okay, cool?

I wouldn’t exactly call this empathy-building tech, but I would hear an argument that it’s an example of tech showing empathy for its users, with the usual caveat that this is probably way too little, way too late. It seems like a good thing, don’t get me wrong. It just should have been a thing much sooner.

This won’t have much use for me, because I’ve already unfollowed or blocked the people whose comments I’d least like to see. What I’d really like is a pop-up kind of like what Netflix has, that alerts me after I’ve been scrolling for more than 15 minutes… “Maybe it’s time for a break?” Or the ability to customize a pop up for when I visit one of my frenemies’ accounts… “Remember why you unfollowed this person??” But I could see it being useful for a teenager who gets bombarded with bullying messages. It’s a start, at least.

Nextdoor, essentially a neighborhood-specific Facebook/Reddit hybrid, did recently release prompts that might encourage empathyLike all social media platforms, Nextdoor has gained a reputation for fostering nastiness, NIMBYism, and even racism. So it launched a “kindness reminder,” which pops up to let you know if your reply to someone’s comment “looks similar to content that’s been reported in the past” and gives you a chance to re-read the community guidelines and rephrase your comment.

Nextdoor says the feature is meant to “encourage positivity across the Nextdoor platform,” but they also seem to suggest that it will make neighborhoods themselves more kind. They claim that in early tests of the feature, 1 in 5 people chose to edit their comments, “resulting in 2-% fewer negative comments” (though it’s not clear to me exactly how they measure negativity). They also claim the Kindness Reminder gets prompted less over time in areas where it’s been tested.

This, like Instagram’s Restricted feature, is an example of a social media company responding to many, many, many complaints of negative behavior and impact. But in Nextdoor’s case, there at least seems to be more transparency. In their post explaining the new feature, Nextdoor says the company built an advisory panel of experts, including Dr. Jennifer Eberhardt, a social scientist who wrote a book about racial bias. There was apparently a session with some of Eberhardt’s students in which Nextdoor employees (executives? unclear) shared their experiences with bias in their own lives as well as on the platform. So, that’s something. If nothing else, I could imagine the Kindness Reminder at least making me stop for a second before dashing off a snarky comment, something that doesn’t happen as much as it used to but is still an unfortunate possibility for me…

One big question about all of this, of course, is why can’t we just use our internal “kindness reminders”? Most of us do have them, after all. But it’s hard when, as Eberhardt notes in the Nextdoor press release: “the problems that we have out in the world and in society make their way online where you’re encouraged to respond quickly and without thinking.” We can create as many empathy-focused tools as we want, but as long as that’s the case, there will always be more work to do.

 

*When I first started writing about this stuff, the concept seemed new to a lot of people and it seemed obvious that the words “ostensibly” or “supposedly” or “hopefully” were implied. Today, not so much, for good reason: a lot of tech that’s advertised as empathetic seems more invasive or manipulative. So, I hope you will trust me when I say I understand that context, and I think about the phrase “empathy-building tech” as having an asterisk most of the time.

Is AOC right about AI?

Conservative Twitter is up in arms today over Rep. Alexandria Ocasio-Cortez saying at an MLK Day event that algorithms are biased. (Of course “bias” has been translated into “racism.”) The general response from the right has been, “What a dumb socialist! Algorithms are run by math. Math can’t be racist!” And from the tech experts on Twitter: “Well, actually….”

I have to put myself in the latter camp. Though I’m not exactly a tech expert, I’ve been researching the impact of technology like AI and algorithms on human well-being for a couple of years now, and the evidence is pretty clear: people have bias, people make algorithms, so algorithms have bias.

When I was a kid, my dad had this new-fangled job as a “computer programmer”. The most vivid and lasting evidence of this vocation was huge stacks of perforated printer paper and dozens upon dozens of floppy disks. But I also remember him saying this phrase enough times to get it stuck in my head: “garbage in, garbage out.” This phrase became popular in the early computer days because it was an easy way to explain what happened when flawed data was put into a machine – the machine spit flawed data out. This was true when my dad was doing…whatever he was doing… and when I was trying to change the look of my MySpace page with rudimentary HTML code. And it’s true with AI, too. (Which is a big reason we need the tech world to focus more on empathy. But I won’t go on that tangent today.)

When I was just starting work on my book, I read Cathy O’Neil’s Weapons of Math Destruction (read it.), which convinced me beyond any remaining doubt that we had a problem. Relying on algorithms to make decisions for us that have little to no oversight and are entirely susceptible to contamination by human bias – conscious or not – is not a liberal anxiety dream. It’s our current reality. It’s just that a lot of us – and I’ll be clear that here I mean a lot of us white and otherwise nonmarginalized people – don’t really notice.

Maybe you still think this is BS. Numbers are numbers, regardless of the intent/mistake/feeling/belief of the person entering them into a computer, you say. This is often hard to get your head around when you see all bias as intentional, I get that, I’ve been there. So let me give you some examples:

There are several studies showing that people with names that don’t “sound white” are often passed up for jobs in favor of more “white-sounding” names. It reportedly happens to women, too. A couple of years ago, Amazon noticed that the algorithm it had created to sift through resumes was biased against women. It had somehow “taught itself that male candidates were preferable.” Amazon tweaked the algorithm, but eventually gave up on it, claiming it might find other ways to skirt neutrality. The algorithm wasn’t doing that with a mind of its own, of course. Machine-learning algorithms, well, learn, but they have to have teachers, whether those teachers are people or gobs of data arranged by people (or by other bots that were programmed by people…). There’s always a person involved, is my point, and people are fallible. And biased. Even unconsciouslyEven IBM admits it. This is a really difficult problem that even the biggest tech companies haven’t yet figured out how to fix. This isn’t about saying “developers are racist/sexist/evil,” it’s about accounting for the fact that all people have biases, and even if we try to set them aside, they can show up in our work. Especially when those of us doing that work happen to be a pretty homogeneous group. One argument for more diversity in tech is that if the humans making the bots are more diverse, the bots will know how to recognize and value more than one kind of person. (Hey, maybe instead of trying to kill us the bots that take over the world will be super woke!)

Another example: In 2015, Google came under fire after a facial recognition app identified several black people as gorillas. There’s no nice way to say that. That’s what happened. The company apologized and tried to fix it, but the best it could do at the time was to remove “gorilla” as an option for the AI. So what happened? Google hasn’t been totally clear on the answer to this, but facial recognition AI works by learning to categorize lots and lots of photos. Technically someone could have trained it to label black people as gorillas, but perhaps more likely is that the folks training the AI in this case simply didn’t consider this potential unintended consequence of letting an imperfect facial recognition bot out into the world. (And, advocates argue, maybe more black folks on the developer team could have prevented this. Maybe.) Last year a spokesperson told Wired: “Image labeling technology is still early and unfortunately it’s nowhere near perfect.” At least Google Photos lets users to report mistakes, but for those who are still skeptical, note: that means even Google acknowledges mistakes are being – and will continue to be – made in this arena.

One last example, because it’s perhaps the most obvious and also maybe the most ridiculous: Microsoft’s Twitter bot, Tay. In 2016, this AI chatbot was unleashed on Twitter, ready to learn how to talk like a millennial and show off Microsoft’s algorithmic skills. But almost as soon as Tay encountered the actual people of Twitter – all of them, not just cutesy millennials speaking in Internet code but also unrepentant trolls and malignant racists – her limitations were put into stark relief. In less than a day, she became a caricature of violent, anti-semitic racist. Some of the tweets seemed to come out of nowhere, but some were thanks to a nifty feature in which people could say “repeat after me” to Tay and she would do just that. (Who ever would have thought that could backfire on Twitter?) Microsoft deleted Tay’s most offensive tweets and eventually made her account private. It was a wild day on the Internet, even for 2016, but it was quickly forgotten. The story bears repeating today, though, because clearly we are still working out the whole bot-human interaction thing.

To close, I’ll just leave you with AOC’s words at the MLK event. See if they still seem dramatic to you.

“Look at – IBM was creating facial recognition technology to target, to do crime profiling. We see over and over again, whether it’s FaceTime, they always have these racial inequities that get translated because algorithms are still made by human beings, and those algorithms are still pegged to those, to basic human assumptions. They’re just automated, and automated assumptions, it’s like if you don’t fix the bias then you’re automating the bias. And that gets even more dangerous.”

(This is the “crime profiling” thing she references, by the way. I’m not sure where the FaceTime thing comes from but I will update this post if/when I get some context on that.)

Update: Thanks to the PLUG newsletter (which I highly recommend) I just came across this fantastic video that does a wonderful job of explaining the issue of AI bias and diversity. It includes a pretty wild example, too. Check it out.