Facebook under fire for burying research into mental health impact

This is an audio transcript of the FT News Briefing podcast episode: Facebook under fire for burying research into mental health impact

Marc Filippino
Good morning from the Financial Times. Today is Friday, October 1st. And this is your FT News Briefing.

[MUSIC PLAYING]

The recent sell-off in government bonds is sending a message about the global economy. We’ll talk about what it’s signalling. And artificial intelligence could improve weather forecasting.

Clive Cookson
Say you’re in a Wimbledon tennis and you want to know when the rain is going to sweep in and stop play or at a cricket match. That can really help scheduling the short time.

Marc Filippino
But first, Facebook. It’s in the hot seat again after revelations that the company knew that its social media sites were harming teens and it didn’t do anything about it. And our European technology correspondent, Madhumita Murgia, says it may be time to turn off Facebook’s digital fire hose of ads. I’m Marc Filippino, and here’s the news you need to start your day.

[MUSIC PLAYING]

Facebook was hammered yesterday by US lawmakers from both political parties. They hauled the company’s global head of safety up to Capitol Hill for questioning. The hearings were called after a Wall Street Journal investigation revealed that Facebook knew of the harm it was doing to teens and other vulnerable users. Here’s Democratic Senator Richard Blumenthal.

Richard Blumenthal
We’re here today because Facebook has shown us, once again, that it is incapable of holding itself accountable. We now have deep insight into Facebook’s relentless campaign to recruit and exploit young users.

Marc Filippino
Also at issue is Facebook’s advertising. It’s the engine of the company’s profits, and it’s Facebook’s advertising algorithms that determine the ads you see on your social media feed. The FT’s Madhumita Murgia looked into this and wrote a column that says, despite the company’s claims that users can control what they see, people really can’t control what ads show up. Madhu joins me now to talk about this. Hey, Madhu.

Madhumita Murgia
Hi, hi, Marc.

Marc Filippino
Madhu, you start your column with a story, really sad story actually, about how these ads can cause a lot of pain for someone going through a tough time. Can you tell us a little bit about it?

Madhumita Murgia
Sure. Really, everybody has, I call it like, an algorithm war story. You know, a good example, I thought was this woman who had miscarried twins, and she was constantly followed by ads about parenting, maternity, you know, related to these things. But the thing that really stood out to me about her was in her attempt to try and stop these ads, you know, she just started to repeatedly Google the word miscarriage because she hoped that the algorithm would pick up on that and stop showing her these ads. And that just really kind of struck me as, you know, there is no way to stop these algorithms. There’s no way to turn it off, really. And I wanted to try and figure out the ways that Facebook does provide to turn off certain types of ads. Doesn’t really work.

Marc Filippino
So Facebook does give users some control over the ads they see. You dug into this a bit, citing research done by a civil liberties organisation in Poland and another researcher at Northeastern University. What did you find?

Madhumita Murgia
Sure. What I discovered is that while there are a range of tools available to help us kind of fine tune what we see, actually, the algorithm still finds a way around it. So all of the tools currently seem to be a way to tell advertisers what we do and don’t want to see. So that but that’s just one way in which you’re targeted on Facebook, right? But, you know, Facebook has its own algorithm. And that algorithm is it operates under its own principles. And those patterns that it spots, those are kind of opaque and invisible even to Facebook itself. And so the algorithm can still figure out things that you like, for example, based on how you behave on Facebook, what ads you click on, what groups you’re a part of, all that sort of thing. And then also it incorporates the knowledge it gains about what you’re doing off of Facebook, like what websites you’re visiting, what you’re buying, what you put in your shopping cart, you know, what videos you’re watching. And together it builds up this picture of you that. based on this research, even if you’ve turned off all these categories and all these topics, it still can figure out that this person is interested in or seems to engage with health-related things or parenting-related things or, you know, cancer-related things. The result is no matter how many of these different tools you have, unless you switch off the algorithm, you’re still going to see ads.

Marc Filippino
Why can’t people just, you know, disengage from Facebook or basically get off the platform entirely?

Madhumita Murgia
Well, I mean, you know, I’ve seen people comment this and say this. And, of course, you can say to people like if you don’t want to be tracked or profiled, whatever, just get off the platforms. And, you know, in many ways, yes, if you stopped using Facebook, Instagram, WhatsApp, everything, you know, you wouldn’t be part of the ecosystem. They might continue to hold all the data about you that they’ve collected previously, but they couldn’t target you going forward, right? My point, really, I came away from it thinking, you know, there shouldn’t be a need for people to completely go off of a platform. They might have a use for that platform. It might be useful in lots of different ways to connect them to people they couldn’t otherwise be connected to. And, you know, really, they should have the option of not being targeted in a way that impacts them negatively. So we’ve got to look at it more broadly, I think, as a society and figure out if we want to continue living on the internet and using these services, do we have to resign ourselves to getting sicker? In the case of teenagers who might become depressed from using it, do we have to just resign ourselves to that? Or can there be a legislative way in which we can say, just turn off the targeting? And, you know, that’s a question really for governments to figure out if they can legislate around that.

Marc Filippino
Madhumita Murgia is the FT’s European technology correspondent. Thanks, Madhu.

Madhumita Murgia
Thank you.

[MUSIC PLAYING]

Marc Filippino
It’s been a rough week for government bonds. In fact, September was the worst month for the global bond market since earlier this year. Now, part of that has to do with concerns about inflation, which markets had shrugged off until this past week.

Katie Martin
And suddenly it sort of dawned on the bond market that,oh central banks are actually serious, and, you know, maybe they’re going to start turning off the taps.

Marc Filippino
That is, of course, our markets editor, Katie Martin. She says bond markets are like a signal of what the global economy is doing next. So what were markets signalling this week?

Katie Martin
This is the market’s way of saying, we think that this funny pick-up in inflation that we started to see at the start of this year might not just be a fleeting little head fake. You know, at first you could write it off as, oh, it’s just used car prices. Oh, there’s lots of little one off effects that come from the rebound from the pandemic. Actually, some of these pricing pressures are proving much stickier than people initially thought. And the strains that you’re seeing in supply chains are just jacking up the price of everything that you and I buy every day. And this is the bond market’s way of reflecting that. Now that the real bogeyman actually is stagflation. It’s where economic growth stagnates, but inflation picks up. Now, to be clear, nobody’s saying that we’re going to have a sort of 1970s redux with double-digit inflation. But what they are saying is this is kind of stagflation, 2021 edition. That’s quite a toxic combination for companies.

Marc Filippino
Now, usually investors pile into bonds when stock prices go down, but this week, stocks went down along with bonds. We asked Katie what she makes of that.

Katie Martin
We’re still seeing a decent performance in stock markets, but we are seeing some of the gains that we got used to at the start of this year and the back end of last year really running out of steam. So it looks like we’re going to get yet another quarter of gains in the S&P 500 so the US benchmark of stocks, but things are just slowing down. And again, that’s a reflection of the market saying, yeah, we think central banks might be forced into doing something here, they might be forced into raising rates or at the very least pulling back some of their bond purchases because inflation is just proving much more persistent than we thought it would. Investors are definitely getting more jumpy. And the assumption is, you know, the easy gains are behind us.

Marc Filippino
Katie Martin is our markets editor.

[MUSIC PLAYING]

And before we go, I want to let you in on a conversation I had with our science editor, Clive Cookson, about the weather.

Clive Cookson
The weather on this side of the Atlantic is particularly variable and hard to predict, I think by global standards.

Marc Filippino
Yes, on the surface, the weather is not the flashiest subject to talk about here on the briefing. But Clive did some really interesting reporting recently about a new tool that might help meteorologists better predict this tricky weather. It involves artificial intelligence. Research by the UK Meteorological Office and London-based AI company DeepMind shows that AI can improve the accuracy of short-term weather forecasts.

Clive Cookson
What this new artificial intelligence research project is doing is going for one of the most difficult short-term predictions of all. And that’s when it’s going to rain, where it’s going to rain and how hard it’s going to rain or snow.

Marc Filippino
And weather forecasting using AI won’t just help us know when to grab an umbrella, it could actually save lives.

Richard Blumenthal
What the Met Office and DeepMind envisage as the key utility will be predicting the most intense rainfall, which is most likely, of course, to cause flooding. That can be very, very hard to predict in other ways. But if you know where there’s going to be an absolute deluge, say, sort of inches of rain falling per hour, you can get people evacuated from low-lying dwellings, basements, get them out before they’re flooded.

Marc Filippino
Clive Cookson is the FT’s science editor.

[MUSIC PLAYING]

You can read more on all of these stories at FT.com. This has been your daily FT News Briefing. And a quick reminder, don’t forget to listen tomorrow to our new FT Weekend podcast. It won’t be in this podcast feed, but you can just search for FT Weekend on your podcast app of choice and subscribe. And be sure to check back here next week for the latest business news. The FT News Briefing is produced by Fiona Symon and me, Marc Filippino. Our editor is Jess Smith. We had help this week from Peter Barber, George Drake Jr, Gavin Kallman and Michael Bruning. Our theme song is by Metaphor Music.

This transcript has been automatically generated. If by any chance there is an error please send the details for a correction to: typo@ft.com. We will do our best to make the amendment as soon as possible.