The algorithms behind social media platforms can change your attitude towards people with different political views in just one day, suggesting that tech companies really can have a strong influence on how we perceive others.
Most social media platforms use algorithms to sort a user’s feed, generally ranking posts by someone’s likelihood of engaging with them and presenting the most engaging first – but this may inadvertently be pushing people apart.
AI-generated writing is now all over the internet. The introduction of automated prose can sometimes change a website’s character, like when once beloved publications get purchased and overhauled into AI content mills. Other times, however, it’s harder to argue that AI really changed anything. For example, look at LinkedIn.
The Microsoft-owned social media site for business professionals has embraced AI, even offering LinkedIn Premium subscribers access to its own in-house AI writing tools that can “rewrite” posts, profiles, and direct messages. The initiative appears to be working: Over 54 percent of longer English-language posts on LinkedIn are likely AI-generated, according to a new analysis shared exclusively with WIRED by the AI detection startup Originality AI. It’s just that the corporate-speak style of AI writing on the platform can be tricky to distinguish from genuine human-penned Thought Leader Blogging.
Originality scanned a sample of 8,795 public LinkedIn posts over 100 words long that were published from January 2018 to October 2024. For the first few years, the use of AI writing tools on LinkedIn was negligible. A major increase then occurred at the beginning of 2023. “The uptick happened when ChatGPT came out,” says Originality CEO Jon Gillham. At that point, Originality found the number of likely AI-generated posts had spiked 189 percent; it has since leveled off.
LinkedIn says it doesn’t track how many posts on the site are written or edited with AI tools. “But we do have robust defenses in place to proactively identify low-quality, and exact or near-exact duplicate content. When we detect such content, we take action to ensure it is not broadly promoted,” says Adam Walkiewicz, LinkedIn’s head of “feed relevance.” “We see AI as a tool that can help with review of a draft or to beat the blank page problem, but the original thoughts and ideas that our members share are what matter.”
LinkedIn is for finding a new job and keeping in touch with former coworkers, which means it’s a relatively staid social media platform. But in recent years, it’s developed its own network of influencers, and is surprisingly popular with Gen Z, including teenagers. Like everywhere else on the internet, people are thirsty for attention on LinkedIn, too, and startups have realized there’s money to be made helping people grow their audiences. There’s a cottage industry of AI LinkedIn comment and post generators to help the career-minded churn out content to dazzle potential bosses or prospective customers. Instead of spending four minutes puzzling over the right tone with which to congratulate an ex-colleague on their promotion, it now takes four seconds to conjure up an algorithmically-generated accolade instead.
But LinkedIn users who spoke to WIRED say that they rely more on general-purpose large language models to cobble their LinkedIn posts together rather than bothering with specialty AI tools. Content writer Adetayo Sogbesan says she uses Anthropic’s Claude to spin up rough drafts of posts she creates on behalf of clients in the tech industry. “Of course, there’s a lot of editing done after,” she says, but the chatbot still “helps me save a lot of time.”
In the real world, we have more than a century of experience figuring out how to share the world with children in order to keep them safe while still allowing adults to engage in adult-only activities, particularly those involving sex, violence, and addictive substances.
In 18th and 19th century America, there were essentially no restrictions on children’s consumption of alcohol. However, following the temperance movement’s efforts to publicize alcohol’s harmful effects on families, women, and children, and after the failed experiment of Prohibition, states took on the responsibility of regulating alcohol. Each state eventually passed laws prohibiting the sale of alcohol to those under a certain age, usually 21. This established the principle that enforcement responsibility falls to the bars, liquor stores, and casinos profiting from alcohol sales. The idea that parents alone should manage their children’s access to alcohol would have struck most people as absurd.
Likewise, it will soon seem absurd that we once allowed children of any age to go everywhere on the internet that adults go, doing everything that adults do, without the knowledge or consent of their parents. The year 2025 will be the one where humanity remembers children are different from adults and that they need protection and age-gating in some parts of the digital world.
The dangers are now undeniable. From the dawn of the internet through to 2024, any child who knew how to lie about their age could open an account on nearly any platform used by adults, except for those that require a credit card. This included hardcore pornography sites such as Pornhub, and the now-defunct site Omegle—where children could video chat with strangers, some of whom were naked masturbating men. It also included social media platforms such as Instagram, Snapchat, and TikTok, all of which are full of content that is wildly inappropriate for children, and all of which incorporate design features that harm children in a variety of ways.
Concern among parents and educators is now widespread.
In 2023, a survey on children’s health conducted by the C.S. Mott Children’s Hospital showed that the issues that most concern parents—ranked above school violence, drugs, and bullying—were the overuse of smartphones, social media and internet safety. Another 2024 survey of school principals showed that they were similarly alarmed by the effect of smartphones on students, with 88 percent stating that they were making children tired and distracted, and 85 percent believing it was amplifying violence and bullying in schools.
No wonder that, in 2023, a major Unesco report considered the overwhelming evidence that excessive phone use was correlated with lower school performance and poorer mental health, and called for the ban of smartphones from schools. In 2024, France, Italy, Finland, and the Netherlands followed through on those recommendations, banning digital devices in classrooms. In the US, the states of Ohio, Indiana, Oklahoma, Virginia, and Florida have also imposed restrictions on smartphone usage in schools, while the US Surgeon General Vivek Murthy called for warning labels for social media platforms. Bipartisan legislation addressing these concerns—the Kids Online Safety Act—has also passed the Senate. This new law would, for instance, force tech companies from targeting kids with personalized algorithms designed to hook them.
In 2025, parents will no longer be alone in tackling this problem. They will be assisted by concerned politicians and by phone-free schools. Social media companies, on the other hand, will finally acknowledge—or be forced to acknowledge by juries and legislatures—that they now own childhood, and they bear at least some responsibility for what they are doing to children.
Time capsules turn up in the most unexpected places online. This one surfaced by design. It’s a YouTube video, dated September 25, 2010. In it, dozens of people packed on a dark dance floor hold their hands up in anticipation of a beat drop. When it does, more hands go in the air. Grainy and less than 60 seconds long, the video is a remnant of that time, three years after the first iPhone, when people were still learning of its capabilities and house music was entering its Coachella bro phase. The video, filename IMG 0107, has nine views.
IMG 0107 landed on my screen by way of IMG_0001, a website created by San Francisco engineer Riley Walz that pulls all the videos uploaded to YouTube from the iPhone’s long-lost “Send to YouTube” feature. Because iPhone used to name video files “IMG_XXXX,” Walz says he was able to use YouTube’s API to pull all the videos with names in that format. He identified about 5 million. On his site, those videos cycle through in no particular order, like a playlist on shuffle, offering up what Walz calls “unedited, pure moments from random lives.” It’s the kind of single-serving site few people make these days, but also one that speaks to the current yearning for a bygone digital era.
“It’s almost like these videos are kind of extinct now,” Walz says when I call to ask him about his site. “They won’t really be produced this way ever again. It’s like a time machine.”
Nostalgia for the lost internet runs rampant in certain corners. Bluesky, which has been gaining about a million users per day since Election Day in the US, is full of people looking to re-create the Twitter of circa 2009, before the platform was awash in slurs and trolls. As WIRED reported earlier this week, fans had to scramble to save Sexypedia’s data after Fandom erased the wiki, taking the internet’s repository of Tumblr Sexymen offline. Tumblr, meanwhile, is always dying. People who want to remember what the internet looked like a decade ago often rely on the Internet Archive’s Wayback Machine, but even the future of that database feels uncertain.
Remembering the internet of yore remains, somewhat ironically, one of the web’s favorite pastimes. Folks still wax poetic about the Space Jam website. (Officially, it’s now a landing page for the LeBron James–fronted 2021 reboot, but the old site still lives at spacejam.com/1996.) Sites like BuzzFeed, which now itself feels old school, still often run listicles of internetmemories. But Internet Archaeology, a site devoted to collecting old home pages, is gone. (WIRED has a small collection of its findings.)
Googling around for this story I was served an AI Overview that informed me that “remembering the old internet” refers to “looking back at the early days of the World Wide Web.” Thanks. I also got an old Reddit feed, a WIRED story, and a piece from The Atlantic about “digital rot”—the phenomenon of the disappearing web that online archivists want to save. The problem with archiving, though, remains that you can archive a static image of an AOL Instant Messenger screen, but you can’t archive the feeling of getting kicked off of chat because your mom picked up the phone. Same goes for the feeling of seeing that a celeb liked your tweet, something most people haven’t felt in a long time.
Leah Feiger: And in the studio with me is WIRED reporter Vittoria Elliott. Hey, Tori.
Vittoria Elliott: Hey, Leah.
Leah Feiger: Guys, how are you doing? What social media platforms are you on right now?
Vittoria Elliott: A violent question if I’m honest. I’m all in on Bluesky.
Leah Feiger: Yeah?
Vittoria Elliott: Oh yeah.
Leah Feiger: Okay.
Vittoria Elliott: I’m just hanging out and people are following me. It’s the lowest lift of any platform I’ve had to engage in in a while.
Leah Feiger: So you just like the attention?
Vittoria Elliott: I do just like the attention. I love getting little pings on my phone every day validating me.
Leah Feiger: It’s what Twitter used to be for so many people. David, you’re on Bluesky too, right?
David Gilbert: I sure am. I’m even following you, Leah, finally.
Leah Feiger: I know. I had to beg for it. It was upsetting.
David Gilbert: I’m very busy though. I have lots of new followers to be vetting and checking.
Leah Feiger: To attend to.
David Gilbert: Yeah.
Leah Feiger: So rude. So what’s up? Do you guys like it more than X right now? There’s just been this massive exodus it feels like to me of not only journalists, but left leaning people across the board. What are the reasons why?
David Gilbert: For me at least it’s just less triggering. It’s less toxic. It’s just easier to be there for a while and scrolling and not be triggered by something stupid that someone has said, which is what X had turned into recently. And I think I’m happier just to spend time there. The engagement seems to be more natural, authentic. Just conversations happening rather than just people shouting all the time. And I don’t know, it’s very early days and I think it could potentially be really good, but it could also potentially be terrible because it’s the internet.
Leah Feiger: The CEO says that they’re gaining like a million users a day. That is a wild stat.
Vittoria Elliott: Yeah. I mean, so I agree with David, the user experience is just better. I mean, even though it’s a little slower in the load because they are gaining so many people so quickly, it is a platform that you feel is designed not to vacuum up all of your attention. So if you notice like with X, sometimes you’ll find a really interesting post and then the website will auto load and then it’ll push you back up to the top, so you have to keep scrolling. And Bluesky doesn’t function that way. It feels like old social media where you’re like, I’m here to see the things I want to see, but this isn’t designed in a way that makes me feel like the ultimate goal is just to make me lose as much time on this website as possible.
Under a Trump administration, the outcomes for 2025 appear inevitably bleaker. He’s promised to invest in an economy that is anti-woke, bolstering his cabinet with agitators—such as Brendan Carr, his choice for chair of the Federal Communications Commission—that have promised to end DEI. Project 2025, the 900-page conservative policy agenda Trump is likely to base much of his governing around, takes aim at organizations that employ “racial classifications and quotas” and pledges to rescind an executive order that calls for federal contractors to guarantee equal opportunity. (And big tech companies were already cutting DEI programs, even without the threats of a hostile president.)
“The idea that DEI is hurting productivity is asinine,” said Maryland governor Wes Moore, the only state representative in attendance. “Look at the numbers.” A 2020 report from McKinsey & Company, for example, shows that diversity, equity and inclusion initiatives are, in fact, good for business. “Authenticity” was the most popular buzzword of the week—vigorously repeated in every discussion I attended— which felt both on brand and eerily odd, given that the business of authenticity will be a target across the next four years.
“We’ve never seen what’s about to come,” artist Will.i.am said onstage, and that much was true.
All the big players were accounted for at the recruiting expo— Netflix, American Express, Axon, Meta, Google, Oracle—as people stood in snaking lines that sometimes stretched longer than those for the night’s afterparties. As I took in the extravaganza of the exhibition floor, with its large swooping signs in every direction, I thought back to my first day in Houston when a Microsoft recruiter joked that I shouldn’t tell anyone what he did, worried they might overwhelm him with resumes and questions about openings at the company.
Still, it was hard to square just how effectively the conference was preparing its next generation of heroes. It wasn’t a question of programming but rather impact. Naturally, all eyes are on AI, but the concerns of others were noticeably elsewhere, in the here and now—and that meant landing a job.
“I’ve always been on edge about job security. I’ve always had uncertainty,” said Candace Madison, who works in legal tech at Relativity, a data organization software company in Chicago. It was her first time at AfroTech. “I don’t think the election increased that, but with the election and DEI not being a priority, you do have to be more on your toes,” she added. Still, she was optimistic. “The way to stay ahead of everything that happens now is networking,” even though she admitted that she’d met very few people in her field so far.
In the elevator at Le Meridien in downtown Houston, a graduate student completing her PhD in data science who was also on the job hunt, colored her experience another way. “This is my eighth [conference] of the year,” she said. “I’m doing my best networking but I’m not getting much out of them.”
On Instagram, the conference was promoted as a success. In a story post, a product engineer at a Fortune 50 company movingly explained how the conference was a “full circle moment” for him, having landed an internship at the expo in 2017 that led to his current job. Another post, from a high-ranking marketing executive, described this year’s experience as “a balm in Gilead.”
As might be expected, everyone at AfroTech had their eyes set on the future—only, no one could say what was coming next, or how much of a say they would have in it. Everyone, it seemed, wanted a piece of what they felt was owed to them: the promise of a stable tomorrow. How they would get there was another matter entirely.
As a technology reporter, I like to think I’m an early adopter. I first signed up to the social network Bluesky around 18 months ago, when the platform saw a small surge in users disaffected by Elon Musk’s approach to what was then still called Twitter.
It didn’t stick. Like many, I found the lure of Twitter too strong, and let my Bluesky account wither, but in recent weeks I have returned – and I am not alone. With Musk continuing to transform his social platform, now called X, at the same time as taking a role in US president-elect Donald Trump’s upcoming government, the Xodus has begun. Bluesky has gained 12 million users in two months, and is fast approaching 20 million users. This time I intend on sticking around – and I think others will, too.
In large part, that’s because I want a social media experience without being bombarded by hate speech, gore and pornographic videos – all of which users of X have complained about in recent months. But I’m also big on Bluesky because I think it signals a shift in how social media works on a more fundamental level.
Social media algorithms – the computer code that decides what each user is shown – have long been a point of contentious debate. Fears of disappearing down “rabbit holes” of radicalisation, or being trapped in “echo chambers” of consensual, sometimes conspiratorial, viewpoints, have dominated scientific literature.
The use of algorithms to filter information has become the norm because chronologically presenting information from followers creates a confusing morass for the average user to process. Sorting and filtering what is important – or likely to keep users engaged – has become key to the success of platforms like Facebook, X and Instagram.
But control of these algorithms also gives you a big say in what people read. One of the bugbears many users have with X is its “For you” algorithm, which under Musk has seen commentary by and about him seemingly shoved into users’ timelines, even if they don’t directly follow him.
Bluesky’s approach isn’t to ditch algorithms – instead, it has more than the average social network. In a 2023 blog post, Jay Graber, Bluesky’s CEO, outlined the ethos of the platform. Bluesky promotes a “marketplace of algorithms”, she wrote, instead of a single “master algorithm”.
In practice, this means that users can see posts by people they follow on the app, the standard view Bluesky defaults to. But they can equally opt to see what’s popular with friends, an algorithmically-dictated selection of posts that your peers enjoy. There are feeds specifically for scientists, curated by those working in the field, or ones to promote Black voices, which are often thinned out by algorithmic filtering. One feed even specifically promotes “quiet posters” – users who post infrequently, and whose views would otherwise be drowned out by those who share every opinion with their followers.
This menu of options allows Bluesky to serve two purposes, bridging the past era of social media and the future one. The platform has the potential, once it reaches a critical mass of users, to act as the “de facto public town square”, as Musk dubbed Twitter before he purchased it. Bluesky arguably is the only remaining such square, given X has shifted to exclude many mainstream voices, and competitors like Threads choose to shy away from promoting politics and current affairs.
But Bluesky also allows you to tailor the app to your needs – not only through feeds, but other elements like starter packs of recommended users to quickly get involved in individual niches, or blocking tools to quieten unruly voices.
There are still hitches, undoubtedly. Finding the right feed for you can be tricky, while creating your own is even more complicated, requiring third-party tools. But the ability to get the full view of public conversation, then to drill down into smaller debates within clusters and communities of that broad swathe of society, is exciting. It’s a model of a new social media where users, not big companies or enigmatic individuals, are in charge of what they see. And if Bluesky continues to add users, it could become the norm. So come and join me – I’m @stokel.bsky.social.
Chris Stokel-Walker is a freelance technology journalist
