AI is Killing VRChat

Credibility score: 48/100 — Mixed Credibility. Several questionable claims detected. Watch with healthy skepticism.

BSmeter analyzed "AI is Killing VRChat" and rated it 48/100 for credibility (a BS score of 52/100 — mixed credibility), on 2026-10-02. Its weakest claim — "Claims you must 'adapt or be left behind' without nuance" — scored 20/100 and was flagged as false dilemma. 41 claims were checked against the video transcript. Scores are produced by BSmeter's AI analysis of the transcript, not independent human verification.

Of 41 claims analyzed: 3 scored under 40, 34 between 40 and 69, and 4 at 70 or above.

Claims analyzed

Uses high-quality human art as a setup to claim AI isn't necessary — No Frame (75/100)

At 0:00

Just setting the stage with actual evidence of human skill. No tricks yet, just showing us what we're fighting for 🛡️

Why this score: The speaker is using visual evidence of human-made assets to establish a baseline before making an argument. It's a direct comparison rather than a rhetorical leap.

Original quote: “Take a look at these. What do you think of them? I'll tell you what I think. I think they're really pretty. These are good worlds. And every one of them is made by people, human beings. I'm prefacing this video with these worlds to make a point. You don't need AI to make this stuff.”

Video opens with a highlight reel preview of AI vs human creativity — Just Vibes (50/100)

At 0:30

Setting the stage with a quick montage of AI-generated assets vs human effort 🎭

Why this score: A rapid sequence of clips meant to bait the viewer into a debate about automation and art.

Original quote: “don't need AI to make this stuff. And when it's all done by hand, it's 100% more impressive. Cuz that's not what I've been seeing recently.”

Claims 'everyone' can't detect AI or doesn't care — Dubious (45/100)

At 1:23

Using 'everyone' to mask a tiny sample size of people he actually talked to 💀

Why this score: He's using the weight of 'everyone' to make a personal anecdote sound like a universal truth.

Original quote: “pretty much everyone I talked to either couldn't tell what was AI generated or really didn't care.”

Claims you must 'adapt or be left behind' without nuance — False Dilemma (20/100)

At 1:41

Reduces the complex reality of AI to a binary choice: adapt or die 💀

Why this score: He's presenting a two-option ultimatum that ignores the nuance of how people actually interact with technology.

Original quote: “AI is here, you know, like no matter what you want to say about it, like you either adapt it or get left behind.”

Claims thumbnails are being fake-upscaled by AI to create clickbait — Loaded Language (45/100)

At 2:48

Using 'AI hell' and 'countless clickbait' to turn a technical observation into an emotional crusade 💀

Why this score: The speaker uses heavy-handed, loaded terms like 'AI hell' to frame a simple observation about image upscaling as a massive cultural problem. It's more attitude than data.

Original quote: “This journey into AI hell starts with thumbnails. They took the photo here and then the AI upscaled it right here from this angle... Countless AI clickbait.”

Claims AI is polluting thumbnails and content without clear distinction — Loaded Language (45/100)

At 4:34

Using 'obviously AI enhanced' to create a sense of inescapable corruption 💀

Why this score: The speaker uses loaded terms like 'obviously' and 'problem' to frame the mere presence of AI as a contamination of the creative process, rather than just a tool.

Original quote: “I don't know what or if anything in them is AI generated, which is part of the problem. We'll get there. The thumbnails are obviously AI enhanced, which you can definitely see in the more recent stuff.”

Comparing AI to 'back rooms' via loaded metaphor 🎭 — Loaded Language (45/100)

At 6:30

Using creepy 'back room' imagery to make a technical point feel like a horror story 💀

Why this score: He's leaning heavily on the 'uncanny valley' vibe to make a point about AI lacking human understanding. It's effective for drama, but it's pure atmosphere over actual mechanics.

Original quote: “It's as if we've stumbled into somewhere in the back rooms. Actually, that's that that's a good comparison. A recreation made by mimicking what humans do, yet not fully understanding us.”

Claiming old worlds are disappearing via emotional framing 🚩 — Emotional Button (45/100)

At 7:22

Using frustration to claim the old community is being buried by new content 💀

Why this score: He's making a sweeping claim about the 'death' of old content without citing actual world-count data or retention stats. It's a vibe, not a metric.

Original quote: “But if you want to find any of those old worlds, aside from the ones that have stayed relevant and popular and active, good [ __ ] luck.”

Predicting VRChat will ignore player feedback 💀 — Confidence Mismatch (45/100)

At 8:01

Making a bold prediction about developer behavior without any evidence of their process 🎭

Why this score: He's acting like he knows how the VRChat team thinks, but it's just a cynical guess based on past bad experiences.

Original quote: “If they're trying things out, it should be for more than a few days with plenty of player feedback, but I I just don't see that happening.”

Compares AI worlds to a menu of 70% feces using loaded language — Loaded Language (45/100)

At 8:47

Using 'human feces' to describe AI assets is pure emotional heavy lifting 💩

Why this score: He's using a visceral, disgusting analogy to make a point about quality control. It's effective for the gut, but it's not data—it's just pure attitude.

Original quote: “That that isn't the case anymore. These slop worlds outpace any of the legit worlds. And with that comes alienation of the user base. I if you look at the menu at a restaurant and 70% of the menu is human feces that you have to eat if you order it, but only a few of those that are [ __ ] are…”

Speculates devs have different views on AI with zero evidence — Confidence Mismatch (45/100)

At 9:54

Speculating on the devs' minds like he's reading their thoughts 🔮

Why this score: He's making a massive assumption about the leadership's philosophy without citing a single developer tweet or official stance. It's just an educated guess disguised as insight.

Original quote: “would be the least that can be done. But I worry I worry that the VR chat devs just don't agree and have completely different views on AI. I don't know. We'll see.”

Claims ChatGPT's Astra model got millions of views via a clip — No Frame (75/100)

At 10:12

Reporting on a massive social media trend—hard to argue with millions of eyeballs 👁️

Original quote: “On Thursday last week, ChatGpt launched their their new model Astra with this clip showing a house it made using Blender. And this racked up millions of views across social media.”

claims GPT6 Astra is driving tool awareness via marketing, using loaded context — Loaded Language (45/100)

At 10:30

Using 'garnered a lot of attention' to make a marketing trend sound like a cultural shift 🚩

Why this score: The speaker is using high-velocity buzzwords like 'GPT6 Astra' and 'garnered attention' to build momentum without explaining the actual mechanism of how these tools are spreading.

Original quote: “because of GPT6 Astra in a marketing campaign that has garnered a lot of attention. With that, a bunch of people were made aware of a tool that exists for Blender and Unity, MCP, model context protocol”

claims users lose fundamental skills by relying on AI tools, using emotional weight — Emotional Button (45/100)

At 11:02

Framing the loss of skill as a crisis for creativity to hit an emotional nerve 💀

Why this score: He's making a heavy-handed argument that using AI is inherently 'destructive' to human understanding, turning a technical workflow into a philosophical tragedy.

Original quote: “Everything here was modeled by the LLM, no third party assets were used at all. ... It stops being a tool real quick. The majority of those who make use of it don't really understand the program that they use it for.”

claims the lack of human input makes the work less acceptable, using loaded framing — Loaded Language (45/100)

At 11:59

Using 'acceptance' and 'influence' to turn a technical process into a moral debate 😈

Why this score: He's using highly subjective terms like 'accepting' and 'influence' to judge the value of a piece of work based solely on its source, rather than its quality.

Original quote: “I'd be more impressed if it wasn't entirely done by an AI. I'd be more accepting if the map wasn't entirely done by an AI. No human influence aside from prompts.”

Claims AI users bypass learning and 'learn nothing' — loaded language 😈 — Loaded Language (45/100)

At 12:32

Using 'learned nothing' to describe anyone using a tool is pure emotional manipulation 💀

Why this score: He's taking the concept of 'efficiency' and rebranding it as 'lack of knowledge.' It's a classic trap: making the shortcut look like a cheat code to make people feel morally inferior for using it.

Original quote: “I've been there. So, people try to bypass it. Bypass that process of learning with AI like this. This guy's learned nothing.”

Claims using AI tools makes you a 'failed creator' — emotional button 🚩 — Emotional Button (45/100)

At 13:09

Calling someone a 'failed creator' just because they used an automated tool is peak drama 🎭

Why this score: He's tying creative identity to the manual labor of fixing technical errors. It’s a massive reach—using tools doesn't mean you failed, it just means you didn't want to deal with the messy technicalities he loves so much.

Original quote: “But they're wrong. It's not hard... If you have Astra make a map for you and you don't know what's wrong with it. You you just you just see what you see and it seems fine at first glance and you roll with it, you failed as a creator.”

Claims AI worlds cause lag and have a 'sickening' vibe — loaded language 💀 — Loaded Language (45/100)

At 13:49

Using 'sickening' to describe a technical lag issue is pure attitude 😈

Why this score: He's turning a technical performance issue (lag) into a moral or aesthetic judgment. It’s not 'sickening,' it's just inefficient code, but he wants you to feel the disgust.

Original quote: “There is a reason why everyone that I've showed these AI worlds to lags out whenever they join. Plus, it gives this off vibe paired with AI thumbnails. It's sickening.”

Calling a game a 'straight rip' using AI to mask it 🚩 — Loaded Language (45/100)

At 14:30

Using 'straight rip' and 'cheese clicker' to build an emotional case for theft 💀

Why this score: The speaker is using highly charged, informal language to frame a creative choice as an act of deception. It's not a technical analysis; it's a moral judgment on the use of AI to bypass copyright.

Original quote: “that's got 800 people in it. It's Cookie Clicker VR. But they changed it. Now it's Cheese Clicker. cheese clicker to avoid what the copyright allegations even though it's this is a rip. This is just a straight rip of Cookie Clicker and then they used AI to change it to cheese. I don't know. I don't…”

Labeling the game as 'AI slop' and outright theft 🚩 — Loaded Language (45/100)

At 14:52

Using 'slop' and 'theft' to turn a technical similarity into an ethical crime 😈

Why this score: The speaker is using aggressive metaphors like 'slop' to characterize the quality and intent of the work, making a subjective observation feel like an objective truth.

Original quote: “This is a prime example of the AI slop that is flooding VR chat. Why? Cuz it's theft. It's literally like a onetoone replication of Cookie Clicker. Cookie Clicker is a free browserbased game. What this person has done is rip it with AI. They've made it playable here. It's It's a rip. One for one.…”

Claiming massive profit from AI 'flips' in the creator economy 💰 — Missing Context (45/100)

At 16:04

Claims he's making 'so much money' without showing a single receipt 💸

Why this score: The speaker asserts significant monetary success from these AI-driven 'flips' but provides no specific data or scale, leaving the actual economic impact vague and unverified.

Original quote: “And worst of all, he's made so much money through the creator economy doing this. VR Chat needs to step up its game when it comes to who they accept into the creator economy because quick AI flips are something I've been seeing more and more and we're only going to see more of it. When there's…”

Uses a Cookie Clicker analogy to frame AI theft as unfair 🍒 — False Equivalence (20/100)

At 16:38

Comparing a unique game design to AI-generated assets is a massive stretch 💀

Why this score: He's equating the theft of intellectual property in software development to the generation of assets via AI, which are fundamentally different levels of 'creation' and 'recreation'.

Original quote: “I mean, think about it this way. You're the creator of Cookie Clicker and then someone recreates your game and is earning money but in VR chat. Is that fair?”

Claims Japanese players are ignoring copyright in AI usage 🚩 — Missing Context (45/100)

At 16:47

He's judging an entire culture's copyright stance based on a specific subculture of VRChat players 🎭

Why this score: He's making a massive leap from the behavior of VRChat users to the legalistic nature of Japanese society, missing the nuance of how AI-assisted creation is viewed in Japan's tech scene.

Original quote: “This is why I find it so confusing that Japan specifically is at the front of the AI movement on VR chat. It's mostly Japanese players. With how vocal and latigious the Japanese side is when it comes to copyright and protecting their works, they seem to not give a [ __ ] when a robot steals instead…”

Uses YouTube search results as proof of universal AI-driven repetition 💀 — unverifiable (50/100)

At 18:03

Using a single YouTube search as 'proof' of a global internet trend is peak amateur hour 📉

Why this score: While the observation about repetitive content is real, using it as a universal 'experiment' to prove an internet-wide problem is more of a personal observation than hard data.

Original quote: “I want to take a step back from VR chat for a second here cuz cuz this is not a VRC exclusive problem. It's a wider internet problem that no matter what sector you can't get away from. For a fun little experiment you can do at home, literally just type in why it sucked into the YouTube search bar…”

Claims AI YouTubers prioritize metrics over quality using 'steal ideas' framing — Loaded Language (45/100)

At 18:55

Uses loaded terms like 'steal' and 'unfiltered greed' to frame a business model as a moral decay 💀

Why this score: The speaker uses highly emotive language to describe the mechanics of content creation, turning simple market research into a cynical critique of 'unoriginality' to trigger an emotional response.

Original quote: “First thing you want to do to make these videos is you need to find an inspiration channel to kind of steal video IDEAS FROM. ... They don't care about making quality content. They care about numbers, printing money, analytics, line go up.”

Claims music is crumbling via a specific stat about Spotify songs — Cherry-Picked (20/100)

At 19:50

Using a single week's snapshot of one platform to claim an entire industry is 'crumbling' 🍒

Why this score: A single data point from one week on a single platform is the definition of cherry-picking to create a sense of catastrophe. It ignores long-term trends or other platforms.

Original quote: “The music scene is crumbling, too. Last week, six of the top 50 trending songs on Spotify were AI generated.”

Speculates Spotify uses AI to drain money from real artists — Confidence Mismatch (45/100)

At 19:59

Uses 'most likely' and 'at least' while sounding like a definitive conspiracy theorist 🎭

Why this score: The speaker is making strong claims about corporate strategy while explicitly using hedging language like 'most likely,' creating a mismatch between the intensity of the tone and the certainty of the evidence.

Original quote: “Drew Gooden made a fantastic video on how Spotify is shoving AI slop in all the popular playlists, most likely made by them, or at least funded by them, to bleed money away from actual artists”

Claims musicians work as ghost-producers for Spotify using pseudonyms — Anonymous Authority (45/100)

At 20:12

References 'several musicians' without naming a single one to back up the claim 🕵️

Why this score: By citing 'several musicians' without specific names or verifiable accounts, the speaker relies on the weight of collective experience rather than hard evidence to validate a very specific industry secret.

Original quote: “Several musicians have since revealed that they used to have a gig with Spotify where they would just record a bunch of like two-minute long instrumentals that then Spotify would upload to their platform under random pseudonym.”

Uses a specific 'Mrs. Hart' anecdote to illustrate AI exploitation — Personal Story (60/100)

At 21:31

Using one person's 'unmasking' to represent a whole industry. It's high drama, but it's her story 🎭

Why this score: The speaker uses a single, highly specific case study—a woman with 21 pen-names and 200 novels—to build a narrative of widespread industry corruption. It's an anecdote used to make the abstract feel personal and threatening.

Original quote: “There's a woman who has been abusing this system and profiting from it and teaching others how to do so. I'm going to call her Mrs. Hart because that was one of her pen names. And over the course of about 8 months, she created 21 different pen names and published over 200 AI generated novels.”

Claims AI growth leads to local pollution and higher bills — Loaded Language (45/100)

At 21:55

Connecting global AI hype to your personal electricity bill. That's a massive leap in logic 💀

Why this score: The speaker uses heavy-hitting words like 'polluting' and 'raising your bills' to create an emotional connection between global tech trends and the viewer's immediate domestic life. It's a massive logical leap to claim AI data centers are the primary driver of an individual's utility bill.

Original quote: “And the result of all of this is governments pushing to build AI data centers in your town, polluting the area, raising your electricity bill.”

Uses a specific 137% stat to claim electronics are unaffordable — Cherry-Picked (45/100)

At 22:18

Using one specific SSD price spike to claim all electronics are 'unaffordable'. That's a massive stretch 🍒

Why this score: The speaker takes one extreme statistical outlier—a 137% increase in a specific type of high-capacity SSD—and uses it to make a sweeping generalization that 'all electronics' are now at 'unaffordable levels.' It ignores the stability of other hardware components to make a more dramatic point.

Original quote: “And that the prices of all your electronics have gone up to unaffordable levels. Since last September, the average price of SSDs and RAM has continued to climb. Prices have increased by 137% for two TB NVME SSDs”

claims US tariffs under USMCA make hardware expensive via political manipulation — Loaded Language (45/100)

At 22:51

Using heavy political drama to explain hardware price hikes. It's a bit much for a discussion about SSD prices 💀

Why this score: The speaker uses high-stakes political framing ('politics will fuck with you') to justify a personal cost increase, making it feel like a grand geopolitical tragedy rather than just market volatility.

Original quote: “Over the past year, the US has imposed a series of tariffs that violate its commitment under our agreement with them in Mexico, so-called Kisma, have made it so damn expensive. It's one of those you might not [ __ ] with politics, but politics will [ __ ] with you”

argues human greed and corporate exploitation are ruining the AI dream — Emotional Button (45/100)

At 23:08

Pivot from 'AI will free us' to 'human greed is a trap.' Pure emotional manipulation of the tech hype 😈

Why this score: The speaker uses a classic 'idealism vs. reality' emotional pivot, relying on the concept of 'human greed' to frame AI integration as a predatory conquest rather than a technical shift.

Original quote: “When we imagined a future world with AI technology in it, we imagined the robots doing the boring tough work and the humans being able to relax and do the things that they're passionate about, art. But in those fantasies, there was a factor that we never considered, human greed. We never considered…”

claims corporate AI adoption is eroding human culture and mental stability — Loaded Language (45/100)

At 23:48

Using terms like 'eroded by slop' to describe cultural shifts. It's heavy on the drama, light on data 💀

Why this score: The speaker uses highly evocative and loaded language ('slop', 'eroded') to turn a corporate trend into an existential crisis, making it feel more like a manifesto than an observation.

Original quote: “The CEOs who want to fire all their staff and have robots do all the work so as to maximize profit at any cost. When you take a look around and you see all of this stuff happening, you see real consequences to your friends and family. You see your favorite things, the things that you consume and…”

frames VRChat as a 'safe place' being threatened by AI — Emotional Button (45/100)

At 24:16

Calling VRChat a 'safe place' to be protected. It's an emotional plea for nostalgia 🚩

Why this score: The speaker frames the platform as a sanctuary to build an emotional connection with the audience, making the threat of AI feel like a loss of 'safe' human-only territory.

Original quote: “You want to protect the small safe places you have left. VR Chat was built up by these creatives, the worlds and avatars made with intense care by incredibly talented people.”

Blaming other platforms' 'pollution' for the fear of AI in VRChat — emotional framing — Loaded Language (45/100)

At 24:32

Using 'polluted' to describe AI integration is pure drama. It’s a vibe, not a metric 💀

Why this score: The speaker uses highly charged language like 'polluted' and '[ ]' to frame the arrival of AI as a dirty, invasive force. It’s an emotional argument based on how other platforms 'feel' rather than actual data on community sentiment or platform stability.

Original quote: “that community that built this game doesn't want to see their corner of the internet polluted by the same [ __ ] that spread elsewhere. It's not necessarily a reaction to the AI worlds themselves, but a reaction brought on by seeing what the other platforms have become when they accept this stuff…”

Claims people are 'glazing' the tool with zero real testing evidence — Confidence Mismatch (45/100)

At 26:43

He's acting like the hype is a fact when he hasn't even shown me the actual community feedback or testing logs 🙄

Why this score: The speaker uses high-confidence language ('that's all you'll find') to describe the online sentiment without providing any actual proof of this 'glazing' behavior.

Original quote: “This is Messy AI. Their new Meshi 6 preview has some of the cleanest AI 3D models I've seen. In fact, if you go online and you try to find anyone testing it or telling you how they work, that's all you'll find. People glazing it so much.”

Exposes affiliate links as 'hidden ads' via a cynical framing — Loaded Language (45/100)

At 27:20

Calling them 'ads without payment' is a pretty aggressive way to describe standard affiliate marketing 💀

Why this score: The speaker uses loaded terms like 'damning' and 'ads without payment' to manipulate the viewer's perception of standard affiliate marketing practices.

Original quote: “And even more damning, when you go into the description of these videos, you find every single one of them has an affiliate link. These are ads without the payment upfront. So, they don't have to list the fact that it is an ad.”

Claims AI services fail to meet advertised quality standards — No Frame (75/100)

At 27:56

Actually showing the blurry textures vs what was promised. I can't even lie, he's got receipts here 🔍

Why this score: The speaker provides direct visual evidence of the discrepancy between advertised quality and actual output, making this a straightforward observation.

Original quote: “The truth is that most of these mesh generation services are not as perfect as they claim. Unlike the advert that showed individual freckles and eyebrows, I got very blurry textures. And part of the reason is because texture resolution isn't everything.”

AI generation is a time-consuming, expensive mess of low quality — Loaded Language (45/100)

At 28:30

Using words like 'crap' and '10x the credits' to turn a technical process into an emotional headache 🙄

Why this score: The speaker is using highly charged, subjective language to describe the inefficiency of AI tools. While it's a valid frustration for creators, calling the output 'crap' is an emotional judgment rather than a technical metric.

Original quote: “You explain what you want, provide some photos, and they [ __ ] it up. So, you change things, pivot, try new stuff, and ultimately spend 10x the credits you meant to get a fairly simple object. Then, the UV, topology, and texture you're given are still crap. So, you need to use other AIs and more…”

AI models are often just repurposed professional files being sold as new — No Frame (75/100)

At 29:43

Calls out the specific trick of using existing files as 'AI-generated' proof. Real shade 💀

Why this score: This is a specific, verifiable observation about how some AI services showcase assets. It exposes the 'magic trick' being played on users by revealing the asset is actually a known, purchasable file like 'Cyberbe'.

Original quote: “The model that they're pretending was generated with Meshi, is actually a professionallymade 3D print file called Cyberbe, which you can buy from numerous 3D marketplaces.”

See the full analysis with timestamps →