news

한어Русский языкEnglishFrançaisIndonesianSanskrit日本語DeutschPortuguêsΕλληνικάespañolItalianoSuomalainenLatina

background

This time Flibustier's joke went too far.

Starting at 19:00 GMT on July 18 (03:00 BJT on July 19), Microsoft Windows systems in many parts of the world experienced "blue screen crashes". Many industries such as aviation, medical, media, finance, retail, and logistics were affected. An independent cybersecurity company called Crowdstrike later admitted that the crash was caused by an antivirus software they launched. The antivirus software, called Falcon, was originally designed to protect Microsoft Windows systems from malicious attacks, but after the latest update, it caused large-scale IT failures. According to Microsoft's estimates, the incident affected 8.5 million Windows devices worldwide.

Shortly after the outage, at 5:39 p.m. Beijing time on July 19, Flibustier posted a post using his eponymous X-platform (formerly Twitter) account, in which he described himself as an employee of Crowdstrike, "uploaded a small update on my first day at Crowdstrike," and attached a selfie with Crowdstrike's company logo as the background.

Screenshot of a post published by Flibustier on platform X (formerly Twitter) at 5:39 pm Beijing time on July 19.

Two hours later, he tweeted again, saying he was "fired," and then uploaded a video saying he was "too excited and released a system update without testing" and "waiting for the dismissal letter."

On July 19, Flibustier posted that he had been fired.

At first, people didn't know why Flibustier did this - more than 45 million people viewed his first tweet about the blue screen incident, and more and more people took screenshots of Flibustier's tweets, calling him the "culprit" of Microsoft's "blue screen incident."

Related online rumors posted screenshots of the incident, calling Flibustier the "real culprit" of the "blue screen incident."

It wasn't until July 20 that Flibustier posted a clarification video, saying it was a joke and that the pictures and videos he posted were all made by generative AI. But he noticed that more and more people were discussing his posts seriously, including journalists and computer scientists.

Screenshot of the clarification video released by Flibustier on July 20.

tool

In fact, Flibustier's joke is not hard to spot. If you look closely at the photo he posted, you will notice that his hand in the victory sign is surprisingly small compared to his body. There is a clear distinction between the edge of the head on the right side of the figure and the background, and there is a clear bulge on the upper right side of the scalp, which is a sign of cutout.

The part marked with a red circle in the picture has obvious traces of post-processing.

The Paper took a screenshot of the Crowdstrike logo on the right side of the image and searched it, and found a picture with a similar background on an office scene design website called Facilitatecorp. After downloading this picture and comparing it with the photo posted by Flibustier, it can be found that the two pictures not only have the same aspect ratio, but also have the same size, and the backgrounds of the two pictures are almost completely overlapped, except for the people in the photo posted by Flibustier. This shows that the picture posted by Flibustier was reprocessed based on the picture uploaded on the Facilitatecorp website.

The Facilitatecorp website uses Crowdstrike as an example, showing a picture with a background similar to the one uploaded by Flibustier.

Comparing the pictures provided by the Facilitatecorp website with the photos released by Flibustier, it can be found that not only do the two pictures have the same aspect ratio, but also at the same size, except for the characters in the photos released by Flibustier, the background parts of the two pictures can almost completely overlap.

In addition, the time when Flibustier posted is also suspicious. As we all know, the "blue screen incident" occurred at 19:00 Greenwich time on the 18th, which is 3:00 Beijing time on the 19th, and Flibustier posted the first post at 17:00 on the 19th. At this time, 14 hours have passed since the incident. If Flibustier is really responsible for the incident, when did he update the software?

Looking at Flibustier's X platform account profile, it can be found that he has never revealed his identity as a programmer. On the platform, he calls himself a "digital citizen, trainer and teacher in critical thinking, specializing in fake news and social networks, OSINT (open source intelligence, editor's note), artificial intelligence."

Profile of Flibustier's X-platform account.

On LinkedIn, you can find an account with the same name as Vincent Flibustier, which is his real name. According to his LinkedIn profile, Flibustier has website development skills and is currently a freelance trainer in Belgium, providing training for 200 teachers each year in digital education, OSINT, artificial intelligence, fake news, social networks, etc. However, there is no experience related to Crowdstrike in his resume.

A screenshot of Flibustier's LinkedIn profile.

In September 2021, France.tv reported on Flibustier, mentioning that he was the founder of a satirical news website called Nordpresse. The Paper has previously introduced the genre of satirical news many times, emphasizing that although it is not strictly "false information", it is often confused with serious news when the context is lost, and it needs to be carefully distinguished.

In September 2021, France Television (france.tv) reported on Flibustier, calling him the founder of the satirical news website Nordpresse.

Combining the above clues, it is not difficult to confirm that Flibustier is not an engineer of Crowdstrike, let alone the "culprit" of this Microsoft blue screen incident.

But the question is, given so many obvious loopholes, why did so many people take Flibustier's joke seriously?

In a clarification post published on July 20, Flibustier mentioned a concept called "confirmation bias." Flibustier believes that his joke can spread because it is funny enough and people are willing to believe it.

According to Encyclopedia Britannica, confirmation bias is the tendency to process information by seeking out or interpreting it in a way that is consistent with one’s existing beliefs. In the 1960s, cognitive psychologist Peter Wason found in experiments that people tend to seek out information that confirms their existing beliefs.

Specifically, "confirmation bias" includes three dimensions, namely "research bias", "interpretation bias" and "memory bias" - "research bias" is also called "attention bias", which makes us selectively focus on information that can confirm our views, while ignoring or belittling data that cannot confirm our views; "interpretation bias" makes us consciously interpret information in a way that confirms our beliefs; "memory bias" makes us selectively remember information that supports our views, while forgetting or ignoring information that does not support our views.

Three types of "confirmation bias". Information from the mental health knowledge website verywellmind.

In recent years, more and more studies have found that "confirmation bias" is closely related to the spread of false information. This is because, in our interaction with "fake news", we tend to agree with information that is consistent with the facts we want, and rarely verify such information. Since we are less able to question what we like, we are easily confused by fake news that is consistent with our ideas.

So, is there any way to overcome "confirmation bias"? It's not easy, but experts still offer some suggestions. For example, verywellmind, a website focusing on mental health knowledge, suggests that in life, we should pay attention to the signs that we may fall into confirmation bias, consider all available evidence, seek different perspectives, and be willing to change our minds based on new evidence. Harvard Business School suggests that in the workplace, in order to avoid falling into confirmation bias, we should learn to ask neutral questions, and even designate members in the team to play the role of "devil's advocate" to create objections.

Some advice from Harvard Business School on how to avoid confirmation bias in the workplace.

In addition to "confirmation bias," Flibustier believes that there are other factors that contributed to the rapid spread of the joke as "false information," including:

1. People want to identify the specific "culprit".

2. The "culprit" appears to be a stupid person who is proud of his stupidity. The character he invented fits this description perfectly - the programmer made a big mistake on his first day at work and took a selfie proudly.

3. People want brand new information, and fake news is inherently new because readers won’t read it anywhere else.

4. “Using English” equals “making it easy to share the message internationally”, while at the same time, the vast majority of people have no idea who is spreading the message.

5. Deliberately revealing a harmless error to distract readers from noticing the hole in the message: Due to improper AI processing, his fingers look strange in the picture, like a baby's fingers, but they distract people from other things (for example, he has a horn on his head due to improper cropping).

6. Ironically, this information was initially pushed as a joke, but as it spread, people gradually believed it as a serious matter.

Screenshot of the factors summarized by Flibustier that contributed to the spread of the joke he created in the form of "false information". The "confirmation bias" mentioned in item 6 has been detailed in the previous article.

The above factors can serve as a reference for us to identify fake news in our daily lives. The Paper will also share more knowledge related to fake information in the future to help readers understand the principles of fake information generation and how to combat it.

The Paper intern Hu Huiwen also contributed to this article.