Views are the most visible metric on the internet. You can see, in more or less real time, how many views something got on YouTube, Instagram, TikTok, Facebook, and most other video platforms. X tracks views for every single thing you post, as does Threads. A view is the universal currency of success — more views, more fun.
Technology
On TikTok, YouTube, X, and everywhere, ‘views’ are lies
But it’s all nonsense. Views are nothing. Views are lies.
You may not need me to remind you of this. We’ve known for years that view counts are meaningless, to the point that Facebook wound up getting sued for aggressively inflating view counts in an effort to convince people to make Facebook videos. Others have written thoughtfully about how stupid view counts are. But we still talk about view counts, view counts are still everywhere, so let’s talk once again about view counts.
A “view,” in reality, is not a universal metric. It’s not really anything. It is whatever a platform wants it to be, which usually has no actual correlation to whether someone actually encountered and experienced a piece of content. You can just make the views whatever you want! And if you don’t like the way the numbers look, make views something else!
Let’s just run through a few of these, shall we? The simplest ones to understand are the social platforms: Instagram, TikTok, and (as of last week) YouTube Shorts all count a view the second a video starts playing. This is objectively absurd. Every time you scroll, even if you immediately jump to the next video, the platform logs that you watched the video the same as if you’d seen the whole thing. That’s like saying, if you’re in a Best Buy and you walk past a TV playing Pirates of the Caribbean: At World’s End, you’ve now technically seen Pirates of the Caribbean: At World’s End. Congratulations, you’re a pirate.
In a way, though, that ridiculously easy bar to clear is actually a more accurate measure than some others. On Facebook, for instance, a view is defined as “the number of times a reel or video was played, plus the number of times photos or text were on screen.” Since videos autoplay all over the platform, those two metrics are effectively the same thing. The metric is so unhelpful that Facebook actually offers creators two other numbers: three-second video views, also known as “people who pressed play,” and one-minute video views, which is at least slightly closer to “people who actually watched this thing.” Those numbers aren’t public, though, because they’d be much lower.
The view has been the universal Meta metric since last fall, when Facebook combined all its other performance and engagement metrics into just one. For photos, text posts, and Stories, the company wrote in a blog post, “Views are calculated as the number of times they appear on a person’s screen, including repeat views.” That used to be a different metric — your content being presented to someone was known as an “impression,” but they had to interact with it in some way before it became a view. Now it’s just views.
The idea that everything in your feed counts as a view is pernicious, and it’s everywhere
The idea that everything in your feed counts as a view is pernicious, and it’s everywhere. As you scroll on X, every single post on your feed gets a view as it flows up and off your screen. Posts that appear in search results, on someone’s profile page — anything that shows up on the page appears to be considered “viewed.” X’s documentation on post views is sketchy and vague, but its video rules are pretty straightforward: if the video was playing for at least two seconds, and half of the player was in view on your screen, then that counts as a view. All these videos play automatically, so we’re back to the same thing: if it loaded, you viewed it.
The reason so many companies have embraced such stupid metrics is both simple and self-reinforcing. If you’re the platform that counts views in a way that actually reflects reality, your numbers will be lower. Creators might see that, decide your platform doesn’t have the juice, and start posting somewhere they’ll ostensibly get more eyeballs. Advertisers might worry that they’ll be broadcasting to dead air. On the social web, momentum is everything, and sometimes you have to lie about the size of your party to get the first people in the door.
In this way of defining views, the platforms also have all the control. Think about it: you don’t press play to get the video going, and you don’t have to stick around for it to count. Whatever the platform wants to get views, gets views. There is no step two, no intermediary, no actual matching of content and audience. There are just… views.
Even the Hollywood types are being pulled into the vortex of made-up view counts. Netflix once clocked a view only after you’d completed 70 percent of something — which, I should point out, is the closest thing to actually tracking whether you’ve watched something of any metric we’ve discussed so far. Now, it only takes two minutes for Netflix to decide you’ve watched something. Netflix actually picked two minutes because it’s “long enough to indicate the choice was intentional.” First of all, no it’s not. Second, Netflix knows how much you actually watched! It just wants the numbers to be higher — around 35 percent higher than under the previous metric, Netflix admitted.
Ironically, Netflix is one of the few streamers that explains how it calculates views at all; most keep their metrics quiet, so they can say things like “it was a huge hit!” without having to provide any actual information. Even YouTube is cagey about its calculations: it’s generally accepted wisdom that you have to watch 30 seconds of a standard YouTube video for it to count as a view, but if that’s official policy I sure can’t find it anywhere.
It is incredibly obvious, by the way, that all the companies peddling these fake numbers know what they’re doing. If they thought public-facing view counts were legit, they’d offer those same numbers to creators and advertisers. Creators typically get to see non-public data like watch time and actual interactions, but even they are consistently being given less and less to work with. Advertisers, though, have the run of the place: YouTube and other platforms still track impressions separately from views, but only for ads. (YouTube may count every Shorts scroll as a view publicly, but it only pays creators for what it calls “Engaged views.”) Many platforms even tell advertisers how many people watched a quarter, half, three-quarters, or all of a video. The platforms themselves are collecting all this data and more, of course, in an effort to better tune the algorithm. They know the answers! But they’ll never show them to you.
We’ve been doing this whole internet thing for a while now, and it’s pretty clear that just about all the metrics are bad. They’ve turned the internet into a game to be won, a system to be gamed, a race to the biggest numbers even when the numbers don’t mean anything. Maybe we’d all be better off without the numbers, but they’re not going anywhere. So all we can do is remember: “views” are not views. Views are lies.
MacOS 27 Golden Gate will usher in a bunch of changes to the Mac when it’s released later this year, with its biggest new features revolving around Siri AI. But for now, using the first developer beta, Siri AI is only offered through a waitlist. So what’s available to try is mostly about how the upcoming operating system looks and feels.
You’re not welcomed with any fanfare when you boot up the macOS 27 developer beta (that’ll probably come later), but there’s reason to celebrate. Jump to the appearance settings, and you find that Apple now has a Liquid Glass slider, allowing users to set the amount of UI transparency in macOS. On one end of the slider, it’s as seethrough as Liquid Glass gets, and on the other end the transparent accents are heavily frosted. Golden Gate starts you in the middle of the slider by default, for just a touch of frosting — perhaps a gentle admission that the original look went too far. You sadly can’t go fully opaque, but this frosted look does greatly reduce the distracting elements of Liquid Glass.
After spending just a short while with Golden Gate, I already prefer the minimum transparency look. I’d crank that slider in the full version and never turn back. For the strongest Liquid Glass haters out there, the Reduce Transparency option is still available in the Accessibility settings, but using it is like taking a hammer to all that glass — introducing lots of harsh gray and black backgrounds to the dock, Menu Bar, and Control Center.
The absolute wins for macOS 27’s design is the return of edge-to-edge sidebars with colorful icons and the increased corner radii of windows across the OS. The former is basically a backtrack to the way sidebars used to look (which looked better and easier to parse, with less wasted space). And the latter is just logical. How on Earth did Apple get so high on its own design supply that it allowed windowed apps to have mismatched corners?
I do have my nitpicks — the new battery icon taken from iOS is less legible (really, I hate it). Also, after Apple finally added the most basic window snapping feature in Sequoia, it hasn’t refined it one bit. Both Tahoe and now Golden Gate are leaving me wanting better and faster tiling controls like Windows 11, as well as the simple ability to rename virtual desktops. But so far, nothing.
Apple says Golden Gate is supposed to feel snappier, with faster search indexing. It’s too early to tell how much of a difference this makes on the MacBook Neo I’m testing it on — especially since dev betas are notoriously buggy and unstable. Using Spotlight search for local files on Golden Gate performed similar to another Neo I had on-hand running macOS 26 Tahoe. And opening apps on both systems side-by-side led to mixed results: Golden Gate opened Lightroom Classic and Slack faster, but Tahoe was faster to open Photoshop and Steam. I hope Apple’s under the hood improvements to memory and CPU usage will really show on the MacBook Neo, which could use all the efficiency it can get, but the jury’s out for now.
There’s still more to come with further beta releases of macOS 27, where we’ll at some point be able to fully test Siri AI, Visual Intelligence, and the revamped Spotlight Search. Last year’s power user-focused Spotlight with clipboard history was a nice improvement, but I’m skeptical that Siri AI being baked into Spotlight will be quite the gamechanger Apple’s billing it as. I’ll keep an open mind and be looking to find out once I’m off the waitlist.
For now, I’m relieved Apple is slightly backpedaling on Liquid Glass. While the look was never quite as bad on the Mac as it was on iOS, it’s a welcome change to be able to turn down these transparencies and get a little closer to the old looks from Sequoia. That and the other bits of UI polish are a nice upgrade on their own. Now, Apple has to show that it can nail all the new AI features, too — I’m eager to see how it fares.
NEWYou can now listen to Fox News articles!
Your phone rings. It’s your son’s voice. Panicked. He says he’s been in a car accident. He hurt someone. He’s about to be arrested. He needs $15,000 wired before the end of the day, and please, don’t tell anyone yet. You’d wire the money. Of course you would. Except it isn’t your son.
It’s a scammer who spent about 10 minutes online, pulled three seconds of audio from a Facebook video your son posted last Christmas, and fed it into an AI voice cloning tool that costs less than a Netflix subscription. The voice that broke your heart wasn’t real. The emergency wasn’t real. But the $15,000 transfer? That would have been.
This is already happening to families right now, in every state. And what most people don’t understand is that the voice clone is actually the easy part. What makes these attacks so devastatingly effective is everything that happens before the call.
Join CyberGuy Live: Lock Down Your Phone in 30 Minutes (Saturday, June 13, 10 am ET)
- Your phone holds your email, passwords, photos, banking apps and personal data. In this free, live online class, Kurt the CyberGuy will walk you step by step through simple phone security fixes you can do in real time. You’ll learn how to improve your privacy settings, spot the latest phone scams, use trusted security tools and walk away with a simple checklist to stay protected. Register here: CyberGuyLive.com
AI CYBERSECURITY RISKS AND DEEPFAKE SCAMS ON THE RISE
Data broker profiles can give scammers phone numbers, relatives’ names and addresses that make AI voice scams more convincing. (Matthias Balk/picture alliance via Getty Images)
The technology has crossed a terrifying threshold.
AI can now clone a person’s voice using as little as three seconds of audio, pulled from a social media video, a voicemail greeting, or a voice message. The technology copies tone, speech patterns, and accents closely enough that many people can’t tell the difference between a real voice and a fake one.
Three seconds. That’s shorter than it took you to read that sentence. AI scams surged 1,210% in 2025, and global AI scam losses could reach $40 billion by 2027. This isn’t a slow-building trend. It’s an explosion.
A new study found that 1 in 4 adults have already experienced an AI voice scam. One in four. That’s your neighbor. Your coworker. Someone in your family. But here’s the thing nobody’s telling you.
The voice clone is the last step, not the first
Every article you’ve read about AI voice cloning focuses on the technology. The scarily realistic audio. The three seconds of audio that’s “all they need.” What those articles miss is the setup that happens before the call. A voice clone is useless without answers to two questions: Whose voice do I clone? And who do I call with it?
To answer both of those questions, scammers don’t need to hack anything. They go to the same places anyone can access right now: data broker websites. Armed with your phone number and personal details from a data broker profile, scammers can call you directly and reference your name, address, or recent transactions to appear legitimate. Here’s the step-by-step process, because you need to know exactly how this works.
Step 1: They find you (or your family member) on a people-search site
A scammer types your name into Spokeo, BeenVerified, or Whitepages. Within seconds, they have:
- Your age and current address
- Your phone number
- The names of your relatives, including your adult children and elderly parents
- Where you all used to live
- Estimated household income.
They didn’t hack anything. They paid a few dollars. Or nothing at all.
Step 2: They identify the right target and the right voice to clone
Once they have your family network mapped, they make a decision: Who’s the most vulnerable person to call? And whose voice will make them act?
5 STEPS TO PROTECT YOUR FINANCES FROM FAMILY SCAMS
Often, the target is an elderly parent. The cloned voice is a grandchild or adult child. That combination of a panicked young voice and an older parent who loves them is the most reliably devastating pairing a scammer can manufacture.
Then they go looking for audio. A Facebook video from Thanksgiving. A YouTube clip of a school play. A TikTok your kid posted last summer. Three seconds is enough. The AI tool replicates pitch, cadence, accent, and emotional inflection.
Step 3: They script the emergency
This is where the call starts to feel personal. Data broker profiles can reveal more than your phone number. Scammers may find relatives’ names, rough ages, your city, your property address and other public record details. Then they use those clues to make the fake emergency sound believable.
Scammers introduce physical excuses, like a broken nose or a bad connection, to cover any slight artifacts in the AI voice, then create maximum urgency. The victim is directed to wire money, send cryptocurrency, or hand cash to a “bail bondsman” courier who arrives at the door.
The call sounds real because it was built on real information. Your mother picks up. She hears her grandchild’s voice, the right name, the right emotional register, the right panic. Her rational brain doesn’t stand a chance.
Cybersecurity researchers have noted that the emotional realism of a cloned voice removes the mental barrier to skepticism. When it sounds like your loved one, your rational defenses tend to shut down.
AI voice cloning scams can use short audio clips and public personal data to make fake family emergencies sound real. (Kurt “CyberGuy” Knutsson)
Real families. Real losses
In one documented case in Florida, a woman lost $15,000 after receiving a call from her “crying daughter.” She withdrew cash and placed it in a box, which a driver came to collect from her house. Another call, and a larger money request, soon followed.
The Trapp family in the San Francisco Bay Area received a frantic call from their “son” saying he’d been in a car accident, injured a pregnant woman, and needed urgent help. The scammers posed not only as the son but also as police, instructing the mother to quickly withdraw $15,000 and hand it to a courier already on the way. The family became suspicious just in time and called their son directly. They were the lucky ones.
Hiya’s Q4 2024 Global Call Threat Report found that one-third of survey respondents across the US, UK, Canada, Germany, France, and Spain encountered deepfake voice fraud in 2024 and 30% of those who encountered it fell victim.
“But I don’t post videos of myself online”
Neither did some of the victims’ families. You don’t need to be the one posting. Your grandchild’s TikTok account, your daughter’s Facebook, your son’s YouTube channel, or any public audio of them is all the scammer needs.
And even if your entire family has locked down social media? The data broker profile built on you, listing your phone number, your relatives’ names, and your address, is still there, still searchable, and still pointing scammers directly at the most vulnerable people in your network.
Here’s an uncomfortable truth: data brokers update their databases constantly. Your information can be pulled from voter registration records, property filings, court documents, marketing surveys, and loyalty programs, none of which require your permission. You likely have a profile on dozens of sites right now that you’ve never seen. You can run a free scan to see exactly how exposed you are. Results usually arrive within an hour.
SPRING CLEAN YOUR DIGITAL FOOTPRINT: WHY RETIREES ARE SCAM TARGETS
Why removing yourself from data brokers disrupts the entire attack chain
The voice clone is only one part of the scam. The targeting makes it work. When you remove your family’s information from data broker sites, you cut off the scammer’s research. They may lose access to your mother’s phone number, your relatives’ names or clues about who lives alone. Without that personal map, it becomes much harder to choose the right target and the right voice to clone.
Data broker profiles might link your mobile number to your home address and your relatives’ names. That makes family scams, now frequently enhanced by AI voice cloning, much easier for criminals to execute.
This is why I recommend using a data removal service. It can automatically send removal requests to hundreds of data broker and people search websites on your behalf. It can also keep monitoring and resubmitting requests when your data reappears. Because it will reappear. That’s how these sites work.
Check out my top picks for data removal services and get a free scan to find out if your personal information is already out on the web by visiting Cyberguy.com
Get a free scan to find out if your personal information is already out on the web: Cyberguy.com/FreeScan
Scammers may clone a loved one’s voice, claim there is an emergency and pressure relatives to send money immediately. (Kurt “CyberGuy” Knutsson)
Five things to do right now
Beyond removing your data, do these things this week:
THE ONE THING THAT COULD PROTECT YOUR PARENTS FROM SCAMMERS
1) Create a family code word
Pick something random, “purple cactus,” “blue kettle,” anything unconnected to your actual life. Every family member agrees: any emergency call requesting money must include this word before anyone acts. Scammers cannot guess it. No data broker sells it.
2) Establish a callback rule
No matter how real a voice sounds, hang up and call the person back at their known number, not the number that called you. Real emergencies can wait two minutes for a callback. Scammers count on the panic preventing exactly this.
3) Lock down family members’ social media
Set profiles to friends only. Limit public videos. The less audio of your family that’s publicly available, the harder voice cloning becomes. Talk to your kids and grandkids about this specifically.
4) Warn your most vulnerable relatives directly
Don’t assume they’ll figure this out. Have a specific, explicit conversation: “If you get a call that sounds like me asking for money, stop. Ask for our code word. Call me back at my number. It might not be me.”
5) Never wire money, use gift cards, or hand cash to a courier based on a phone call alone
This is how every one of these scams ends. The payment method itself is the red flag. Legitimate emergencies don’t require Venmo, wire transfers, or a courier showing up at your door.
Kurt’s key takeaways
AI voice scams work because they sound personal. A scammer may only need a few seconds of public audio to copy a loved one’s voice and make a fake emergency feel real. However, the voice clone is only part of the attack. Scammers also use data broker and people-search sites to find phone numbers, family connections and personal details that make the call more convincing. That is why a simple family code word can help stop panic before money changes hands. So can a strict callback rule, locked-down social media and direct conversations with older relatives before a scammer calls. The best defense is to slow the moment down. Hang up, call your loved one directly and never send money, crypto, gift cards or cash to a courier based only on a phone call.
CLICK HERE TO DOWNLOAD THE FOX NEWS APP
If a phone call sounded exactly like someone you love asking for help, would you stop long enough to question it? Let us know by writing to us at Cyberguy.com
Sign up for my FREE CyberGuy Report
- Get my best tech tips, urgent security alerts and exclusive deals delivered straight to your inbox.
- For simple, real-world ways to spot scams early and stay protected, visit CyberGuy.com – trusted by millions who watch CyberGuy on TV daily.
- Plus, you’ll get instant access to my Ultimate Scam Survival Guide free when you join.
Copyright 2026 CyberGuy.com. All rights reserved.
Nearly a year after it was announced, Instagram says it’s delivering the ability to rearrange the posts in your profile grid. It had been available to some people in test groups, but as of June 8th, it’s rolling out widely via the Android and iPhone mobile apps.
Until now, the posts on your Instagram profile have been locked in chronological order beyond the ability to pin three posts at the top, but once the feature is live on your account, you can long-press and drag posts freely, no matter how old they are. Any posts that are pinned will remain at the top.
-
Washington, D.C4 minutes agoTrayon White censured and fined for missing financial disclosure deadlines
-
Cleveland, OH7 minutes agoNew York Yankees vs. Cleveland Guardians prediction, pick for Tuesday 6/9/26
-
Austin, TX12 minutes agoTexas rule targeting smokable hemp is back in effect
-
Alabama19 minutes agoAlabama Football Recruiting Update: Crimson Tide Picks Up Second RB Commit
-
Alaska22 minutes agoThe other Dan Sullivan in Alaska’s Senate race says candidacy is not a sham
-
Arizona27 minutes ago
Arizona police make arrest around Nancy Guthrie’s home
-
Arkansas34 minutes agoSouthern Arkansas University launches new Bachelor of Science in Emergency Management for Fall 2026 | News | Southern Arkansas University
-
California37 minutes agoCalifornia insurance commissioner race is set: Kim vs. Allen
