Donate today and get a Marketplace mug -- perfect for all your liquid assets! Donate now
Why are fake obituaries cluttering Google — and upsetting loved ones?
Apr 4, 2024

Why are fake obituaries cluttering Google — and upsetting loved ones?

HTML EMBED:
COPY
Sites that churn out spam death notices exploit tragedy for profit and have raised concerns about the reliability of search engines. Mia Sato of The Verge investigated the situation.

Obituary spam has been flooding search engines. Among other effects, the disturbing trend has provoked Google to modify its policies.

Marketplace’s Lily Jamali discussed the situation with reporter Mia Sato, who has written about it for The Verge. Her investigation uncovered a network of websites generating this content using search engine optimization, or SEO, tactics. Sato also covered the story of Brian Vastag, a journalist who experienced this abuse when he read his own fake obituary along with that of his ex-wife, who did actually pass away.

The following is an edited transcript of their conversation.

Mia Sato: Brian and Beth were a married couple at one point and then remained very close friends throughout Beth’s life. And in late December 2023, some friends of theirs got news that both of them had suddenly died. Brian was perfectly fine. He’s alive and well. I spoke to him for this piece; he is certainly not dead. Beth had passed away in December. But online, there was a flood of inaccurate news articles and obituaries that claimed both Brian and Beth had passed away. And it became this really doubly traumatizing, doubly upsetting situation where people were mourning for a person who had passed away but then mourning for someone who was very much alive.

Lily Jamali: And what did Brian tell you about how this affected him? 

Sato: He was really upset. He was angry that his friends had had to go through this experience and had been under the impression that he had passed away because it created additional harm. There were [fake obituary] articles written in poor taste or inaccurate about Beth. So, that alone is upsetting enough. But the fact that there was this double impact on his loved ones, I think, made him really upset.

Jamali: You write that these articles don’t contain much information, but they are filled with keywords that target people who might be searching on Google. Why is this happening?

Sato: The sites are the perfect distillation of what it looks like to create content, specifically to rank on Google Search. This is an entire cottage industry, not just for obituaries. Most publishers that rely on Google for traffic will tweak their pages and websites to surface at the top of Google searches. I think another really important part of this is that this is happening to people who have no public profile whatsoever. It’s happening to children; it’s happening to teens and college students. It’s not really about how many people are searching for this. It’s that there’s a rise in people searching for this person’s name, and [these sites] try to jump on that to rise to the top of search results to get some of this traffic and make money. 

Jamali: Well, speaking of that, how do they make money? Because I think that’s key to the motivation here. 

Sato: Another commonality between these sites is that they are loaded with ads. Every time people click on these links or the ad, the website can earn a couple of cents. So, it’s literally pennies, but imagine doing this thousands of times a month for all the different people around the world who are passing away and who people are desperately trying to search for information about. That does add up.

Jamali: From what you wrote, it sounds like [artificial intelligence] has put this industry into a whole new stratosphere. 

Sato: The introduction of AI tools in the last year and a half or two years has accelerated it. Across several sites that I looked at, there were telltale signs that AI tools were being used to produce this content — for example, this discrepancy between claiming that Brian was dead when he wasn’t. We were looking at the sites reporting this, and one website seemed to have taken an op-ed co-authored by Beth and Brian. It appeared as if they just ran it through ChatGPT and requested a summary of the story. There was another website that publishes hundreds of stories a day. Some of the writers published 20 stories alone in a day, which is way too many. But when I looked at the author page, three people were very clearly AI-generated, with headshots with overly smooth faces, unreal backgrounds and weird things happening in their clothes. And after I reached out to the site, they deleted those authors and all their archived stories. The articles all also sound the same. They use unnatural, stilted language and words that you wouldn’t associate with someone actually writing an obituary about someone they knew. This was seen across many different sites. 

Jamali: Obituary scraping has been an issue for quite some time now, and Google has had to deal with it. What are they doing about it?

Sato: So initially when my story was published, Google gave me a sort of boilerplate response: “We know this is an issue, and we are working on it.” Recently, in the last couple of weeks, Google has announced a bunch of algorithm updates for search that it says will cut back on different practices, including obituary spam, which was one of the ones they specifically named. And the idea is that they will cut down on how these sites are able to appear at the top. The responses that I got to the story were people saying: “My friend died last year, and I had this exact experience” or “When my brother passed away, we spent weeks trying to fight these sites and getting these sites taken down or deranked from Google, and it didn’t work.” I think that shows the scale of this problem. Anyone could have someone they love pass away, and they could then have a bunch of websites they’ve never heard of writing incorrect information about them. 

Jamali: You say these are folks who don’t have a public profile. But in some respects, almost all of us have some web profile through social media, if nothing else, right?

Sato: That is for sure true. I think what makes these folks different is that this problem of data voids is a problem for all search engines. The idea behind data voids is that when you search for something, there aren’t enough results to meet that query, so things that shouldn’t be at the top start to bubble up. So, if you search for the name of a person with a Facebook account but no LinkedIn and doesn’t have a public-facing job, and they pass away, what else will Google serve them? They have to put something in front of the users. That is how these sorts of ickier websites can sneak through and slip to the top. Just because they’re at the top at one point doesn’t mean that will be forever. The search rankings do shuffle around. Brian wrote Beth’s actual obituary several weeks after she had passed away, and the last I checked, it was towards the top. So, in this case, he was able to out-SEO the SEO spam sites.

The future of this podcast starts with you.

Every day, the “Marketplace Tech” team demystifies the digital economy with stories that explore more than just Big Tech. We’re committed to covering topics that matter to you and the world around us, diving deep into how technology intersects with climate change, inequity, and disinformation.

As part of a nonprofit newsroom, we’re counting on listeners like you to keep this public service paywall-free and available to all.

Support “Marketplace Tech” in any amount today and become a partner in our mission.

The team

Daisy Palacios Senior Producer
Daniel Shin Producer
Jesús Alvarado Associate Producer
Rosie Hughes Assistant Producer