Launched in 2004 by then-Harvard University student Mark Zuckerberg, Facebook became the biggest social media platform in the world just five years later. It was young people who helped Facebook initially shoot to success, but today’s youth think that “Facebook is for old people”, reported the Daily Star.

In 2015, a survey found that 71 percent of teens aged 13 to 17 used Facebook. But in 2022, that number dropped to 32 percent, reported NBC News. Alex Tani, a 25-year-old from the UK, said that when he started university in 2016, his tutors always set up student groups on Facebook to help everyone stay in touch and interact. But by the time his younger brother went to university three years later, this wasn’t the case.

Today’s younger generation prefers to use other social platforms, such as Instagram, a photo and video-sharing social networking service also owned by Facebook’s parent company Meta, and TikTok, owned by Chinese company ByteDance, said NBC News. These two platforms are different from Facebook in that they are mainly image-based and video-based, not text-based. Visual culture is a rising trend and young people have grown tired of reading people’s remarks on Facebook, said Forbes.

Another reason for Facebook’s decline in popularity is AI. According to The Guardian, TikTok’s “most powerful tool” is its algorithm (算法) which predicts what content you want to see. It shares related content with you from people you may not follow based on what you’ve watched before. This saves people the effort of searching for what they like, and so becomes an “escape for them”, said North Carolina State University.

Going forward, Facebook plans to use AI in a similar way as TikTok to show short videos and give users a better platform to discuss video content, said Fortune Media. Only time will tell whether it can stay “young” or remain as a platform used only by “old people”.

Overhear any conversation or pay close attention to your own, and you will hear laughter. You will probably also notice that, more often than not, the laughter is in response to something that was not very funny—or was not funny at all. Take Hillary Clinton’s strategic laughter during heated exchanges with Donald Trump in the presidential debates.

In one research in my Vocal Communication Lab at UCLA, we played recorded laughs to listeners and asked them, “Is this laugh ‘real’ or ‘fake’?” Our recorded laughs were either taken from real conversations between friends in a laboratory setting, or produced by request, also in the lab. It turned out that about 70 per cent of the listeners were able to tell the “real” laughs from the “fake” ones. Quite a few fake laughs sound pretty good, but why are fake laughs not as convincing as we expect? We can detect part of a laugh that is hard to fake—and that part has to do with the control of our breath.

Laughs could essentially be broken up into two different vocalisation (发声) systems: speech and emotion. Although both kinds of laughs are vocalised as “ha-ha-ha,” the intermittent breaths a person takes in—between the sounds give away a fake laugh. Real laughs have a higher proportion of breathy parts in between. Think about every time you force a laugh. You have to actually say the “ha-ha-ha,” and probably never consider how you breathe while doing it.

A fake laugh is basically an imitation of a real laugh. If you slow down a real laugh about two and a half times, it sounds like an ape (猿). But a slowed fake laugh sounds more like human speech. We put this observation to the test with the slowed-down versions of the laughs used for our first experiment, and asked participants whether the recordings were from a human or an animal. Though they couldn’t tell the origin of the real laugh, they were able to tell that the fake laugh came from a human.

Real laughs are produced by an emotional vocal system that humans share with all primates (灵长目动物), whereas fake laughs are produced by a speech system that is unique to humans. Laughter in humans likely evolved from play vocalisations in our primate ancestors. If one animal bites another during a rough-and-tumble (小打小闹的) play, it could be taken as an attack. But if they signal while panting (喘气) that they are just playing, the play can continue without being interrupted by an unnecessary real fight. In this way, real laughter reveals our animal nature.

Those of us who shop online may scan customer reviews to get a better sense of products we can't judge for ourselves at a physical store. We may check out online testimonials before booking a haircut or visiting a new restaurant. But what happens if some of those reviews can't be trusted?

The Federal Trade Commission (FTC) recently announced that it was proposing new measures to crack down on fake reviews and other practices used to mislead consumers. The commission published a proposed rule that would prohibit companies from writing or selling fake reviews, buying positive reviews, suppressing negative reviews and more.

“Our proposed rule on fake reviews shows that we're using all available means to attack deceptive advertising in the digital age,” Samuel Levine, director of the FTC's Bureau of Consumer Protection, said in a statement.

Research shows people overwhelmingly consult online reviews, but humans are also bad at telling which consumer reviews hold water and which are nonsense.

That's potentially worrisome given that nearly one in every three reviews is fake, according to one estimate.

In 2022, for example, the commission forced the online retailer Fashion Nova to pay $4.2 million to settle allegations (指控) that it blocked negative reviews from being posted on its website — the first FTC action involving a company's effort to hide negative reviews. For its part, Fashion Nova said in a statement to The New York Times that the FTC's allegations were “inaccurate” and that it only settled the charges to avoid “the distraction and legal fees”.

Comments on the proposed rule have to be received within 60 days of its publication in the Federal Register, after which the FTC will decide whether to issue a revised final version.