China is set to establish a new virtual identity system that will allow its citizens to register for several online services and platforms. This new system, currently voluntary, is called the “National Network Identity Authentication.” However, it’ll eliminate one of the few remaining avenues for anonymity for Chinese Internet users.

Virtual ID. The Chinese government published the regulations for this centralized Internet identity system in May 2025, but it won’t come into effect until July 15. Identity checks will become mandatory on all online platforms, making it extremely difficult for users to maintain anonymity on social media and other sites. This virtual ID card will function as a mobile app on NFC-enabled phones and can be set up by both adult and minor users.

Justification. According to official documents, the initiative aims to “protect citizens’ identity information, and support the healthy and orderly development of the digital economy.”

A “bullet-proof vest.” A report by CCTV, China’s official public television service, says that these measures are based on legal identity information and will provide a means to verify users across different services. According to the report, this will reduce the collection and retention of personal identity information. Local media have likened it to a “bullet-proof vest for personal information.”

Digital totalitarianism. Xiao Qiang is a researcher specializing in Internet censorship. He told CNN, “This is a state-led, unified identity system capable of real-time monitoring and blocking of users. It can directly erase voices it doesn’t like from the internet, so it’s more than just a surveillance tool–it is an infrastructure of digital totalitarianism.”

More from Xataka On

Italy Wants to Annihilate the Fake Online Reviews Battering the Restaurant Industry. Its Plan: Banning Anonymous Comments

Yet another repressive measure. There’s the Internet that we all know and use, and then there’s the Chinese Internet. In China, censorship and control pervade every aspect of life, but it’s more evident than ever in terms of Internet access and usage. This has been particularly noticeable in recent years with the widespread ban on feminist and LGBT activism. Youth slang has also been eliminated on social media, while the government has cracked down on celebrity worship. The Chinese regime also has complete control over user comments on social media, allowing it to remove those deemed inappropriate.

State pseudonyms. According to the state media agency Xinhua, more than six million people have voluntarily registered to obtain the new virtual ID card. These “cyberspace IDs” can take two forms: a combination of letters and numbers or an online credential. Both options correspond to the user’s real identity but don’t include plain text information. In other words, they act as pseudonyms, but ones that are controlled by the state.

“Voluntary.” A representative from the Ministry of Public Security told Xinhua that this virtual ID card is voluntary. However, this claim is questionable. The Chinese government is promoting the integration of this system across various industries and sectors. Although the law presents the system as voluntary, it could ultimately become a requirement for users to access different platforms that adopt it, according to Haochen Sun, a law professor at the University of Hong Kong.

Silencing criticism. The system began to take shape over a year ago, when police executive Jia Xiaoliang proposed it during a meeting of the National People’s Congress in March 2024. Widespread criticism from experts followed, with Lao Dongyan, a prominent law professor at Tsinghua University, commenting on Weibo that the system could enable the installation of a “surveillance device on every individual’s online activity.”

Blowing off steam. The post was subsequently deleted, and the user’s account was suspended for three months for “violating relevant rules.” A year later, when the final rules of the project were published, there was virtually no criticism on social media. According to Xiao, this is a common tactic in China, allowing a period for some individuals to “blow off steam” before definitive measures are announced. This tactic diminishes the power of those dissenting voices.

Image | Bo Zhang

Related | China’s Future in the Chip Industry Rests on a Single, Almost Unknown Company: SiCarrier

Internet access failed across North Korea for several hours this weekend. At first, it seemed reasonable to suspect a cyberattack from abroad—possibly retaliation for recent hacks by Lazarus, North Korea’s elite hacker group, targeting foreign institutions and companies. But according to Reuters, several experts say the disruption likely came from inside the country, not from an external cyberattack.

UK researcher Junade Ali monitors North Korean network behavior. Martyn Williams, a North Korea technology and infrastructure expert at the Stimson Center in Washington, said the outage appeared internal, especially since both China- and Russia-based connections were affected. The country’s main news services, the Ministry of Foreign Affairs, and national airline Air Koryo were among the websites that went dark for several hours.

North Korean Citizens Don’t Have Free Access to the Internet

North Korea has one of the most restricted Internet infrastructures in the world. The government strictly controls it to preserve the country’s isolation and ensure that the Kim Jong Un regime maintains full control over what citizens can access. As a result, North Koreans experience a far more limited version of the Internet than users in most other countries.

Still, some institutions inside North Korea have full or near-complete Internet access. Though heavily supervised, government-backed hackers, political elites, military officials, state agencies, and scientists reportedly use a much broader range of online resources than the general population.

More from Xataka On

North Korea Just Launched Its Impressive Frigate With a Vertical Missile System Into the Water. It Sunk

It’s worth noting that North Korea connects to the global Internet through just two fiber-optic links: one with China, managed by China Unicom, and another with Russia.

Perhaps the most surprising aspect for outsiders is that ordinary citizens don’t use the Internet at all. Instead, they connect to a closed national intranet called Kwangmyong, which means “bright light.” Predictably, Kwangmyong is isolated from the global Internet and features only government-approved content.

Available resources include state news, propaganda, educational websites, encyclopedias, and an internal email system—and little else.

Image | Steve Barker (Unsplash)

Related | North Korea Shares Its View on the U.S. Golden Dome: Space Nuclear Warfare Is Closer Than Ever

Nostalgia is an emotional enigma that often eludes explanation. People yearn for past experiences and romanticize them. They compare them with the present, even if they know that those times might not have been as good as they remember. Some miss the early Internet, but wasn’t it filled with chaotic designs, overwhelming aesthetics, and a lack of usability? Perhaps what they truly miss is the untamed wildness of that era.

Web 1.0. The first phase of the Internet as we know it today, known as Web 1.0, spans the period from around 1991 to 2004. Those who lived through it can easily recognize its defining characteristics. Notably, communication was one-way. Users could consume information, but couldn’t interact meaningfully with websites. The content consisted primarily of text and simple images. Designs relied on tables, frames, and basic HTML, and there were no social media platforms or global collaboration. Essentially, the Internet served primarily as a container for information.

Aesthetic. This era had a specific aesthetic that many who lived through it remember vividly, associating it with a particular time and place. Neocities, a hosting service launched in 2013, aims to revive the spirit of Geocities by offering free hosting for websites. It pays homage to that hypertext Internet, which now seems to be fading away. As of now, Neocities hosts over a million websites, with most limited to a maximum size of 1 GB.

More from Xataka On

Chat.com Is One of the Oldest Domains on the Internet. Its New Owner Is None Other Than OpenAI

HTML. Neocities strikes a nice balance between modern features and retro charm. For example, it offers an HTML editor integrated into the browser, and it can also be used to create CSS and JavaScript pages. Free users can only upload files in these three languages, along with Markdown, XML, text, and images. This limitation helps prevent the creation of MP3 repositories, which would ironically fit well with the Web 1.0 era.

You can browse using the website gallery. Organized by tags, it looks like the webrings of the past.

The good old days. What exactly do people miss about the early days of the Internet? That sense of novelty is what probably stands out the most. Every moment online felt like a discovery. However, alongside that discovery was a sense that the Internet was still untamed. The content was wild and unbalanced, reflecting the creators’ interests and desires rather than being shaped by algorithms that dictated what people wanted to see.

Effort. This nostalgia celebrates the “effort” intrinsic to any pioneering technology. The accidental experimental designs and slow connections highlighted the necessity of building the Internet as it evolved. It’s also reflected in the various detours to contact someone, whether through buried email addresses, guestbooks, forums, or early chat systems.

Neocities is both a nod to and a tribute to those times, although with websites that are easier to navigate. After all, the Internet is continuously evolving.

Images | Kimberly Nguyen | Neocities

Related | Digital Serendipity Is at Risk of Disappearing. The Internet Knows You All Too Well

In a way, the Internet is like a double-edged sword, capable of causing great good or great harm in a shockingly short amount of time. Just ask the Marubo Tribe, a 2,000-member Indigenous group in the Brazilian Amazon.

Last year, the tribe was the subject of countless news stories around the world falsely accusing its people of becoming addicted to porn after being given Internet access. According to the Marubo, the spread of this false narrative is all The New York Times’ fault—and it’s suing.

The story. The tribe was the subject of a 2024 story by the Times, which profiled the group’s first exposure to the Internet via Elon Musk’s Starlink, along with the technology’s benefits and challenges. For instance, the story says that while the Internet has allowed the Marubo to video chat with loved ones who live far away and call for help during emergencies, it’s also affected the youth.

“Young people have gotten lazy because of the internet,” Tsainama Marubo, a 73-year-old woman, said. She explained that the tribe’s younger members are less interested in traditional activities, such as making jewelry out of snail shells. “They’re learning the ways of the white people.”

The Internet affected the Marubo in other ways, too—many of them familiar to people who have had access to the technology for decades. Seeing teenagers glued to their phones and group chats filled with gossip became common. In addition, the Marubo started using social media networks and were exposed to misinformation. 

The porn. The Internet also brought the tribe access to porn. The Times story says that minors were now watching porn, which had unsettled some in the tribe. It quoted Alfredo Marubo, leader of a Marubo association of villages and the biggest critic of the tribe’s new access to the Internet.

According to Times, Alfredo Marubo said that young men were sharing mature adult content in group chats. The spreading of that type of content marked a strong contrast with the tribe’s culture, which disapproves of kissing in public.

“We’re worried young people are going to want to try it,” Alfredo Marubo said, referring to the graphic sex shown in the videos being shared.

Alfredo Marubo added that some tribal leaders had told him of young men displaying more aggressive sexual behavior.

The distortion. Overall, based on my count, the references to porn and sex in the story were limited to five sentences. Nowhere did the Times story call the Marubo people porn addicts or suggest it was a huge problem.

That wasn’t the story heard round the world, though. Instead, countless media outlets ran pieces about how the Internet had caused the tribe to become addicted to porn, a claim that was untrue. TMZ ran a story with the headline, “Elon Musk’s Starlink Hookup Leaves A Remote Tribe Addicted To Porn.”

The New York Post also had a headline in the same vein: “Remote Amazon tribe finally connects to internet — only to wind up hooked on porn, social media.”

A little more than week later, the “porn addict” framing had become so widespread that the Times published a follow-up story. In the story, the outlet said that more than 100 websites worldwide had published the false claim.

“The Marubo people are not addicted to pornography. There was no hint of this in the forest, and there was no suggestion of it in The New York Times’s article,” Jack Micas, the author of the piece, wrote.

More from Xataka On

Texas Says It Wants to Ban AI-Generated Child Porn. It May End Up Banning Popular Anime Shows Instead

The lawsuit. But in the eyes of the Marubo people, the damage had already been done. This week, the tribe filed a defamation lawsuit against the Times, TMZ, and Yahoo News in Los Angeles Superior Court, Courthouse News reported.

The suit says that TMZ and Yahoo News are also defendants because their stories amplified and misrepresented the original. The tribe is seeking $180 million from each of the media outlets.

“The NYT portrayed the Marubo people as a community unable to handle basic exposure to the internet, highlighting allegations that their youth had become consumed by pornography shortly after receiving access," the lawsuit said. "These statements were not only inflammatory but conveyed to the average reader that the Marubo people had descended into moral and social decline as a direct result of internet access.”

It added that the stories turned the Marubo into a subject of “international ridicule” and reduced them to “memes and headlines.”

In a statement to Courthouse News, the Times defended the original story and said it did not infer that “any members of the tribe were addicted to pornography.”

“Any fair reading of this piece shows a sensitive and nuanced exploration of the benefits and complications of new technology in a remote Indigenous village with a proud history and preserved culture,” the outlet's spokesperson said. “We intend to vigorously defend against the lawsuit.”

Images | Natalia Pedraza

Related | For 60,000 Years, a Tribe Has Been Isolated on Their Island. A New Threat Has Put Them in Danger: Influencers

Each time you connect to Internet Roadtrip, you appear somewhere new in the U.S. At the time of writing, I’m in Topsham, Maine. Internet Roadtrip uses Google Street View’s vast map to organize a collaborative online journey—frame by frame.

Here’s how Internet Roadtrip works. It’s simple. Log on to the game’s website and access the vehicle’s controls. You and everyone else online—usually between 700 and 900 co-drivers—take turns steering. Every nine seconds, the site asks you to vote on which direction to go next. The most popular choice wins. And then it repeats. Right now, it’s not too exciting—the road in Maine seems endless, and the group can only go one direction. But the journey will get more interesting. Cities, crossroads, and open landscapes are the real highlights.

“Radio Ga Ga.” Even actions like honking the horn or changing the radio station require a vote. (Although, if you chat with other users, you’ll learn there’s an unspoken rule not to skip stations.) The car plays real local radio stations based on its virtual location. That kind of attention to detail—and the slow, democratic pace—has made the site the Internet’s most soothing shared experience right now.

More from Xataka On

One Gorilla vs. 100 Men: The Viral Debate Taking Over the Internet

The Internet belongs to everyone. The road trip began last Tuesday in Boston and hasn’t made it out of the neighboring state. Its creator, Neal Agarwal, told 404 Media he took inspiration from another collective experiment: “Twitch Plays Pokémon,” which has run for a decade. In that game, players vote on every move in a live Twitch stream of Pokémon, resulting in chaotic, unpredictable gameplay. Internet Roadtrip, by contrast, follows Google’s logic. No one’s ending up in a virtual ditch.

Some precedent. This is not the Internet’s first journey across the country. In 2018, writer and photographer Matthew Muspratt used only Street View to travel the U.S. from Rwanda. He crossed 16 states and both coasts. Before that, Wired reported on a couple who made the trip from San Francisco to New York in 90 hours, clicking as fast as possible while carefully planning their route. Muspratt took a more reflective approach, pausing to admire the scenery.

Getting more sophisticated. Virtual travel surged during the pandemic, but projects like Internet Roadtrip now exist purely for fun. And technology keeps expanding what’s possible. Google’s historical Street View lets users jump back up to 80 years in time. Its AI tools can generate personalized routes from photos. But Agarwal has a bigger vision: making collaborative trips possible in a real, remotely controlled self-driving car. That’s still in the future—but maybe not too far.

Image | Internet Roadtrip

Related | Millions of Videos on YouTube Have the Name ‘IMG_XXXX.’ Someone Saved Them, and the Result Is Wonderful

Everyone on the Internet wonders who would win a fight between a gorilla and 100 men. It’s a mental puzzle with no correct answer, meant to spark discussion. That’s why it has become the meme of the moment. Users are approaching it from different angles, from biology to ethics. As they say, the answer matters less than the path to get there.

A war for everyone. The meme is simple: debate who would win in a battle under these conditions. The question went viral in late April. TikTokers like @tredouglass, @lov3charlee, and @rationalsniper have all weighed in (mostly in favor of the gorilla). The dilemma became a global phenomenon when Tesla CEO Elon Musk and influencers like MrBeast tweeted about it.

Origins of the meme. In February 2022, TikTok user @yuri5kpt2 posed the question, generating more than 3,000 comments. Previously, in 2020, Reddit users discussed the scenario in a thread that caught the attention of some media outlets.

Legendary confrontations. Hypothetical battles between mismatched forces are nothing new. Classical military theory often explores asymmetry between combatants—think David vs. Goliath. More recently, games like Age of Empires and Ultimate Epic Battle Simulator have turned such ideas into fantasy warfare, with simulations generating millions of views on YouTube.

More from Xataka On

Digital Serendipity Is at Risk of Disappearing. The Internet Knows You All Too Well

Some are hilariously excessive:

• 100 Teutonic knights vs. 100 monks 
• 100 elephants vs. 100 monks
• 100 modern soldiers vs. 20,000 Spartans
• 1,000 imperial soldiers vs. 450,000 zombie women
• 20,000 giant spiders vs. 120,000 Nazi zombies

Now, the meme has taken on a life of its own, with street interviews, animations, AI experiments, simulations, and countless opinion videos.

Why do people love this debate? We like this meme because we like to argue. As social creatures, we enjoy taking sides—even when the stakes are imaginary. The meme also touches on deeper themes: brute, individual force vs. coordinated, strategic effort. Nature vs. civilization. Memes help us explore big ideas in simple, funny ways.

Evolutionary psychologists told Rolling Out that these kinds of hypothetical debates are modern expressions of ancient survival planning behaviors. In other words, we talk through how we’d face a gorilla—just in case.

Yes, but... who would win? Most experts agree: In a straight fight, the gorilla would win easily. But humans are toolmakers. We’d probably find a way to win, even if it meant improvising weapons. A gorilla is four to 10 times stronger than a person, making it nearly unstoppable in hand-to-hand combat.

Still, the real fun is in the argument—not the outcome.

Image | Amy Reed (Unsplash) | Foto de Bao Menglong en Unsplash

Related | Chill Guy Is the Internet’s Latest Viral Meme Because It’s Something More Profound: Everything We Aspire to Be

First, the power went out. Then mobile networks connection dropped, and people couldn’t make calls. Something big had happened.

Starlink was the exception. With no Wi-Fi or cellular service, many people in Spain were completely cut off. While some operators managed to partially restore service using backup cell tower generators, only Starlink users who could power their routers and antennas enjoyed a stable, fast connection throughout the day.

With more than 7,000 Starlink satellites in orbit, SpaceX’s constellation over the Iberian Peninsula operated without significant issues. The satellites communicate with each other, and while they rely on ground stations to access the Internet, those stations come equipped with generators, solar panels and batteries to handle power outages.

Click on the image to view the original post on X.

A YouTuber in Málaga stayed online during the blackout. Alex, a YouTuber known as Redskull, lives in the Axarquía region of Málaga. He lost power between 12:30 and 17:45 but didn’t regain mobile connectivity until the following day. Starlink, however, kept him online.

“It worked for the entire blackout duration,” he told Xataka On. “Download and upload speeds were even better than usual: about 300 Mbps download and 30 Mbps upload.”

The improved performance likely came from fewer users being online during the outage. “Normally, Starlink speeds are slower at night because the network is more congested. That wasn’t the case yesterday,” Alex added.

More from Xataka On

Currently, All U.S. Phones Can Connect to Starlink Satellites for Free. This Will Change in July

He powered his setup with the Jackery Explorer 1000 V2 Power Station, a large external battery that gave him 12 hours of use, including enough juice to run a mesh Wi-Fi network in his home. Thanks to a 100W solar panel, he didn’t use up the full battery during the day.

Click on the image to view the original post on X.

A connected family in Barcelona. In L’Ametlla del Vallès, a village in the province of Barcelona, the Graells family also stayed connected. Their detached house runs on solar panels and Tesla batteries, and two members of the household work remotely. They learned of the outage only through a notification from the Tesla app—because their Internet stayed up via Starlink.

“The connection was the same as every day, very good for individual use,” Eduardo Graells told Xataka On. “Ping of 22 milliseconds. Download about 150 to 200 Mbps. Upload about 15 Mbps. We used it continuously from early morning until 6 p.m. At 8 p.m., we were watching YouTube and some shows. The power came back in our area at 2:30 a.m.”

Despite the 14-hour outage, the Graells family experienced a normal day, keeping an eye on their battery levels. Their Tesla battery dropped from 70% to 40% during the blackout.

“Starlink can have a one- to two-second interruption when it switches satellites, and it’s very occasional. You only notice it during a video conference. If you’re watching movies, soccer, Formula 1, etc., you don’t notice it—probably due to buffering,” he explained.

Five ground stations in Spain. Starlink recently opened two more ground stations on the Iberian Peninsula: one in the province of Barcelona and another in Santa Olalla, Toledo. That brings the total to five, adding to existing stations in Madrid, Huelva, and Alicante. These facilities proved resilient during the blackout.

The stations are relatively small but easy to recognize by their many phased array antennas, housed in white domes. They connect the satellite constellation to Starlink’s control centers and act as gateways, transmitting client data to the wider Internet. In effect, they serve as critical intermediaries, keeping Starlink’s customers online even when the rest of the grid falters.

Image | Evgeny Opanasenko (Unsplash)

Related | Starlink Satellites Have Revolutionized Modern Warfare. China and Russia Are Developing ‘Starlink Killers’ to Disable Them

In recent years, services like Google Fi have been rising in popularity, with some people ditching traditional operators to switch over to Google’s offering. While Google Fi is a great product for certain people, it’s not the best choice for everyone.

For people considering making the switch to Google Fi from a major carrier like AT&T, here’s what you need to know.

Overview

First off, Google Fi is a MVNO, which stands for mobile virtual network operator. Companies that operate as MVNOs don’t own their own infrastructure, instead choosing to lease the signal, coverage, and network of one of the big three carriers: T-Mobile, AT&T and Verizon. This allows them to offer their services at a lower cost. Other popular MVNOs include Cricket and Mint Mobile.

AT&T is one of the biggest carriers in the U.S. As such, it offers extensive coverage and a variety of plans and features, though it is generally more expensive overall.

Prices and Plans

Google Fi offers three plans. Its entry-level plan starts at $18, though the price can quickly go up if you use a lot of data. The plans all include hotspot tethering, and the more expensive plans feature add-ons like free YouTube Premium, Google One, and shareable data with tablets.

• Flexible: $18 for one person + pay $10 per GB you use. Slower data after 15 GB
• Simply Unlimited: $40 for one person / Unlimited data, throttling begins at 35 GB
• Unlimited Plus: $65 for one person / Unlimited data, slower data begins at 50 GB

Unlike Google Fi, AT&T offers a variety of plans. The company also lets you mix and match its plans, so it’s a little harder to compare them side-by-side. In general, its plans fall into two categories: Unlimited and others. AT&T’s Unlimited plans nearly all offer true unlimited data with no throttling. They also offer at least 5GB of hotspot data per line per month.

While AT&T is great for people looking for family plans, contracting the service as a single person can get expensive. However, the company does offer discounts for teachers, members of the military, and doctors.

Features

One of Google Fi’s most popular features is its international connectivity. Fi offers free data, calls, and texts on its plans within the U.S., Canada, and Mexico. Its Unlimited Plus plan offers free roaming at over 200 destinations.

As for AT&T, I included its true unlimited data above, but it bears mentioning here, too. This is especially important to consider if you or someone in your household watch a lot of video or live events. Most people don’t need unlimited data, but it’s great to know that your speed won’t get throttled after you reach a certain limit.

Coverage

AT&T is one of the big three carriers, so you shouldn’t have problems with coverage or reception anywhere in the U.S. The company offers LTE and 5G coverage, which will ensure you have high-speed Internet.

Although Google Fi leases its network, that doesn’t mean your speed will suffer. Google mainly relies on T-Mobile for its coverage, and the fellow big three carrier gives Fi customers access to its 5G. It’s important to note that during times of high congestion, carriers like T-Mobile and AT&T are likely to prioritize their own customers instead of those on MVNOs.

More from Xataka On

How to Transfer SMS Text Messages to a New Phone

Bottom Line

If you’re single, looking to spend the least money as possible, and don’t use a lot of data, Google Fi is probably the best choice for you. It’ll provide you with everything you need and give you the best bang for your buck.

There’s one caveat, though. If you have an iPhone, you’ll lose out on some functionalities on the Google Fi network, but you’ll need to set up SMS and MMS manually and say goodbye to end-to-end encryption on calls. (You won’t have this problem on AT&T).

On the other hand, if you’re shopping for a family plan and want a good deal for the best service, AT&T is probably the way to go. It’ll give you unlimited data, good hotspot support, and perhaps even a discount if you work in certain sectors. Notably, because you can mix-and-match its plans, you can also end up with a more personalized solution for your needs.

Image | Xataka On

Related | What Type of Phone Do I Have? How to Find Out What Kind of Smartphone Model You Have in Settings on Android and iPhone

In 1995, Yahoo! received its first round of external funding. Within just one year, the tech company increased its advertising revenue from $1.4 million to $19.1 million, representing a growth of 1,200%. The business was rapidly expanding, and by late 1995, Reuters and Softbank had invested an additional $5 million.

In April 1996, Yahoo! went public for the first time, with its stock peaking at $43 on the first day of trading. By 2000, Yahoo!’s stock reached an all-time high of $450 a share. However, just a year and a half later, in September 2001, it closed at an all-time low of $8.11. In a short time, Yahoo experienced a dramatic rise and fall in its value, exemplifying the dot-com bubble phenomenon.

The dot-com bubble rattled stock markets and affected the finances of millions of people. Many argue that a similar scenario is currently taking place in the artificial intelligence market. Since the emergence of ChatGPT, the AI sector has generated immense expectations. In fact, these expectations are even greater than those surrounding the Internet in the late 1990s.

This situation has sparked discussions about a potential AI bubble, with several apparent indicators:

1. Extraordinary valuations, minimal revenues: Many AI startups, including companies like OpenAI, have staggering valuations. However, their revenues are meager compared to their losses. In other words, they’re burning through cash at an alarming rate.
2. Too much competition: Many of these companies are adopting subscription models with several pricing plans. However, the competition is fierce. No single model stands out as superior, leading to the possibility of one winner and many losers.
3. Expectations vs. reality: Executives at these companies are instilling high hopes about AI’s impact on our world, but tangible results are scarce.
4. AI everywhere: The term “AI” is currently overused, diluting its meaning. It seems that every product must incorporate AI, even if it isn’t central to its functionality.
5. Regulatory and ethical challenges: The rapid advancement of AI is also confronting significant regulatory scrutiny. Concerns regarding potential misuse of the technology (like deepfakes) and its negative effects (such as impacts on employment) present barriers to its development.

More from Xataka On

We Finally Know How Thirsty Artificial Intelligence Is: It Needs a Small Bottle of Water to Send a 100-Word Email

The dot-com bubble followed similar patterns, and there are concerning signs that something unusual is happening in the market. Investor behavior appears more speculative than rational, and media coverage is rapidly increasing. Additionally, there’s an overconfidence regarding the future, with risks being largely overlooked. Is the AI bubble a reality?

From the Dot-com Boom to the Potential AI Bubble

There are key differences between the dot-com bubble and AI’s situation today. For one, AI is driving a product that’s just as robust, if not more so, than the Internet was when the dot-com bubble burst. The impact of AI is already being felt in several industries, like programming.

In many cases, AI doesn’t replace human labor but enhances it. AI serves as a new “bicycle for the mind,” helping users of all fields to accomplish tasks more efficiently and effectively.

Unlike the dot-com era, the current push for AI is largely driven by major tech companies, who are determined not to be surpassed by competitors.

Back then, Internet startups such as Yahoo! and Google surprised established giants. Today, while some startups like OpenAI have achieved extraordinary valuations, they face significant competition from well-established companies. Examples include Microsoft, Alphabet, and Nvidia, all of which have benefited from the rising interest in AI.

The potential of AI is as apparent, if not clearer, than the potential of the Internet two decades ago. However, many startups during the dot-com bubble lacked a clear business model. Often, purchasing a .com domain and presenting a compelling sales pitch was enough to secure immediate funding. Investors didn’t want to miss out on emerging opportunities.

The Internet ultimately proved to have an unprecedented impact on humanity, leading to the creation of tech giants that built on that revolution. AI is generating massive and, at times, unsettling expectations. However, whether it’s experiencing another bubble similar to the dot-com boom remains unclear. Only time will tell.

Image | Nicholas Cappello

Related | Microsoft Needs Time to Balance the Revenue From AI Systems Revenue With Their Immense Costs. The Problem: China Is Already in the Picture

Serendipity is described as that valuable discovery that occurs while you’re looking for something else. Unfortunately, it’s on the verge of extinction when it comes to the Internet. Recommendation algorithms are becoming increasingly precise and have trapped users in convenient yet sterile bubbles. We no longer “surf” the Internet as we once did, and that’s the problem.

I think back to my teenage years in the early 2000s. One night, while listening to the radio, I heard something I hadn’t heard before. It wasn’t melancholic commercial pop nor was it the first reggaeton hits. It was Beck’s E-Pro.

Those four minutes changed my perception of music. Today, it may not seem extraordinary. However, at that moment, it sparked in me a desire to explore a genre I had never encountered before. It was an accident, a fortuitous collision with something I would never have actively sought out because I didn’t even know it existed. It simply reached my ears.

As I use Spotify, which suggests songs precisely tailored to my declared and inferred tastes, I wonder where those transformative accidents are for today’s teenagers.

It’s a paradox. The more sophisticated technology becomes at “knowing” us, the less opportunity we have to discover anything truly new. Our algorithms confuse relevance with familiarity, presenting users with barely perceptible variations of what they already consume. Platforms today are designed to hold our attention rather than to expand our horizons.

When was the last time you stumbled upon something genuinely unexpected in your feed? Not just something related to your usual interests, but something entirely new and jarring that made you rethink your ideas and broaden your tastes?

The digital explorers of the past, who navigated from hyperlink to hyperlink, have been replaced by passive consumers swiping through an endless stream of precalculated content. In their quest for refinement, algorithms have eliminated friction, which in turn has stifled the generative spark of disagreement. This isn’t good news.

Ironically, this algorithmic refinement comes at a time when we need divergent thinking the most. Real innovation alters paradigms rather than merely optimizing what already exists. It arises from unexpected connections and the collision of disparate ideas.

Silicon Valley was built on serendipity. Writer Stewart Brand found inspiration in Native American culture to create the Whole Earth Catalog, while Apple founder Steve Jobs was captivated by calligraphy, which profoundly influenced the design and essence of the Mac. Even the concept of hypertext emerged from the analogy of how the human mind works: not linearly, but through unexpected associations.

More from Xataka On

Google Flexes Its Muscles, Unveiling an Impressive AI Agent That Can Browse the Internet for You

It’s not just a matter of innovation. It’s also a matter of civic health. In the past, physical newspapers forced readers to turn pages and encounter opinions that were often discordant with their own. Now, platforms like Google’s Discover do the filtering for us.

Today, our feeds are so finely tuned that you can go months without encountering an idea that truly challenges your convictions. Algorithms, in their eagerness to maximize your engagement, only present what confirms your existing beliefs (or what incites you the most in the case of platforms like X).

This overspecialization of digital consumption has led to a peculiar phenomenon. Despite having unprecedented access to information, our mental worlds are becoming increasingly narrow. Variety is sacrificed for a personalized experience. It’s telling that some of the most influential figures behind these technologies choose not to let their own children use them.

We’re moving toward an Internet where every click is premeditated, where the next recommendation is reliably interesting. In the name of efficiency, we’re sacrificing the delightful digital clutter. Like junk DNA in our genome, digital clutter could contain the seed of the next great innovation or discoveries that might profoundly influence the way we consume culture for the rest of our lives.

I wonder how many groundbreaking concepts in music–like Beck’s innovative tracks–today’s teenagers are missing out on. They’re trapped in algorithmically perfect but creatively sterile content loops.

Perhaps it’s time to demand the right to digital serendipity. We should ask whether we want an Internet that knows us all too well or one that can still surprise us.

Image | Sergey Zolkin

Related | Introverts of the World, We’re Taking Back the Internet