Muse Privacy Breach, Banks Ban AI Agents, and more Singularity News
Credibility score: 59/100 — Mixed Credibility. Several questionable claims detected. Watch with healthy skepticism.
BSmeter analyzed "Muse Privacy Breach, Banks Ban AI Agents, and more Singularity News" and rated it 59/100 for credibility (a BS score of 41/100 — mixed credibility), on 2026-11-08. Its weakest claim — "Claims massive hype for $399 robot ducks via 'reported' pre-orders" — scored 45/100 and was flagged as confidence mismatch. 15 claims were checked against the video transcript. Scores are produced by BSmeter's AI analysis of the transcript, not independent human verification.
Of 15 claims analyzed: 0 scored under 40, 9 between 40 and 69, and 6 at 70 or above.
Claims analyzed
Rapid-fire highlight reel of AI news and tech feats — Just Vibes (50/100)
At 0:00
A chaotic montage of unrelated tech hype to hook you before the real meat 🎢
Why this score: It's a standard teaser loop—jumping from burglaring robots to flying gliders to robot pets. It's not making a claim; it's just selling the chaos.
Original quote: “Lots of meta muse news today, but also robots [music] can now fend off burglars or beat you up. A 26-year-old in China built the Green Goblins glider, then flew it across the Yangze River. The Genie 4 world model shows us a further glimpse [music] at a future of interactive, dynamically generated…”
Claims a major Muse privacy breach was found by experts 🚩 — No Frame (75/100)
At 0:32
Straight facts on the Muse leak. No fluff, just the damage report 💀
Why this score: The speaker provides specific numbers (6,800 chats) and names the firm (Clerk), which aligns with real-world security reports of the incident.
Original quote: “The security firm Clerk uncovered the largest privacy vulnerability in Muse's short history. More than 6,800 shared chats and artifacts were sitting in public DNS records for anyone to find. Fortunately, it was a security firm that discovered this vulnerability instead of any black hat actors. And…”
Claims major Canadian banks are banning AI agents 🚩 — No Frame (75/100)
At 1:31
Naming specific banks and dates. No games being played here 🛡️
Why this score: The claim is highly specific, citing RBC's October 27th move and naming other major banks like TD and Scotia, which is verifiable through news reports.
Original quote: “Canada's biggest banks just drew a line against AI agents. The Royal Bank of Canada blocked third party personal AI agents on October 27th and TD and Scotia Bank are restricting them too. And they named Muse explicitly.”
Claims banks don't trust agents with money-moving power 🚩 — No Frame (70/100)
At 1:47
Explaining the logic behind the ban. It's a cautious move, not a trick 🏦
Why this score: The speaker is explaining the 'why' behind the banking industry's hesitation, focusing on the risk of autonomous transfers.
Original quote: “Basically, they decided we don't trust these AI agents coming in and acting in our customers names, able to make transfers or make withdrawals, anything like that. Plaid readonly access still works like it does before.”
Claims users can 'hack' their way into features by reinstalling the app — Confidence Mismatch (50/100)
At 2:34
Selling a 'reinstall trick' as if it's a legitimate feature access strategy 💀
Why this score: He's presenting user-reported 'hacks' as if they are part of the product experience, without any actual confirmation from Meta that this is intended behavior.
Original quote: “I have had some people on free plans having access, however, and some people have said that if you uninstall the app and reinstall it, it might give you access to the Muse Voice Mode.”
Listing specific regions where Muse is available — No Frame (75/100)
At 3:23
Just stating the rollout map. No tricks here, just geography 🗺️
Why this score: He's providing a straightforward update on availability, though community feedback suggests his claim about Mexico might be slightly ahead of the actual rollout.
Original quote: “Meta is gradually expanding Muse access country by country. Japan and Singapore now have access to Muse on top of North America, which includes Canada, the United States, and Mexico.”
Claims a crowded market is 'good' for consumers — Dubious (50/100)
At 4:05
Calling a crowded market 'good' is just his take. Pure sentiment 🙄
Why this score: He's making a subjective judgment call on market saturation. It's not a fact, just his optimistic take on competition.
Original quote: “But with that on top of Muse, Dots, Grockbot, and Instinct, the AI agent assistant space is getting a little crowded, but that's good for the consumer.”
Claims massive hype for $399 robot ducks via 'reported' pre-orders — Confidence Mismatch (45/100)
At 5:37
Leaning hard on 'reportedly' to mask the hype behind those 20,000 units 🦆
Why this score: The speaker uses 'reportedly' to distance themselves from the validity of the 20,000 unit figure, creating a sense of hype without actually verifying the sales data themselves.
Original quote: “Each $399 robot pet has a unique duck voice and can perform tricks, learn new skills, even rollerblade. They are an open-source development platform aimed at students and hobbyists, and pre-orders reportedly passed 20,000 units after the demo.”
Sources: Nvidia's $399 Microduck Robot Duck Surpasses 20,000 Preorders After Viral GTC Berlin Keynote — BigGo Finance, Nvidia's CEO Played With a $399 Robot Duck and Now It's the Toy Everyone Wants - Teck Hustlers, Jensen Huang's $399 "Duck Robot" Takes Off... 20,000 Preorders After One Somersault - The Asia Business Daily
Unveiling a robot with an absurdly dramatic name — Just Vibes (50/100)
At 6:11
Mocking the 'Kung Fu Legend' branding—it's pure marketing theater 💀
Why this score: This is a commentary on the branding rather than a factual claim, reacting to the absurdity of the name 'H2X Kung Fu Legend'.
Original quote: “Unit's newest humanoid, the H2X Kung Fu Legend, Guardian. What a name. Unveiled October 21st.”
Unitry touts endurance stats while hiding price and autonomy details — Missing Context (45/100)
At 6:44
Selling the 'triple endurance' hype while hiding the price tag and how much is actually autonomous 🕵️
Why this score: The speaker calls out the missing variables—price, release date, and actual autonomy levels—that would make those performance claims meaningful.
Original quote: “Unitry claims roughly 8 hours of endurance, triple the base H2, and pitches it for security patrol and emergency response. Two caveats, no price or release date, and Unitry has not said how much of the video was autonomous.”
A developer used AI to purge bloatware from a Xiaomi TV — Personal Story (60/100)
At 7:08
A tech-bro legend about using Claude to kill annoying TV ads 📺
Why this score: This is a specific user story about using an AI agent to solve a very modern, relatable annoyance with hardware bloatware.
Original quote: “He's a developer got so annoyed that his parents' TV was was showing him ads before he could open Netflix and there was so much bloat wear on the damn thing that he handed the whole machine to Claude Code.”
The project scaled fast but carries the risk of bricking TVs — No Frame (75/100)
At 7:47
Gives the success stats but doesn't forget to mention you might break your TV 💀
Why this score: The speaker provides the scale of adoption while immediately balancing it with a real-world risk/warning from the project itself.
Original quote: “And within days, owners had logged working setups for over 100 TV models from dozens of brands. Fair warning from the project itself, at least one person has reported a bricked television.”
A 14-year-old dropout taught himself engineering to build a glider — Loaded Language (50/100)
At 8:23
Using the 'dropout at 14' narrative to build hype around a DIY engineer 🚀
Why this score: The framing relies on the dramatic 'left school at 14' trope to build a legend around the creator's technical prowess.
Original quote: “Song Jing Yu left school at 14 and taught himself engineering. His first version used five miniature jet”
Comparing a DIY flight prototype to kids on ebikes — loaded comparison 😈 — Loaded Language (45/100)
At 8:57
Using the 'ebike' anxiety to manufacture a sense of impending chaos 💀
Why this score: He's leveraging a common parental anxiety about ebikes to frame a niche engineering project as a societal warning sign. It's emotional framing for the sake of drama.
Original quote: “And he's telling fans not to copy him. Man, if you thought that all the kids having ebikes was bad, just wait.”
Recommending a specific song as a personal preference 🎵 — No Frame (75/100)
At 9:19
Just a guy sharing his playlist. No tricks here, just vibes 🎧
Why this score: This is a straightforward personal recommendation. No attempt to manipulate the listener's opinion on music theory or culture.
Original quote: “I'll end with a recommendation for a slop core song. I've really been enjoying the song White Pill by Anabology.”
See the full analysis with sources and timestamps →