…those who have seen Mr. Bean fighting the insect probably will smile, the maker of this video, @mrabujoe, described it as “a peaceful summer breakfast becomes a full-scale war in a comedic micro-epic about pride, territory, and one very disrespectful insect”…
➔ Grandma vs Wasp (by @mrabujoe on Higgsfield AI > Mr. Abu Joe)
…it’s amazing how such complex videos now can be done with AI, I’ve seen some unbelievable trailers and short movies completely done with AI, that’s why there’s a contest worth $500.000 in prizes on Higgsfield AI, there have been more than 8000 submissions (8751 in total), right now winners are being selected (will be announced in 2 days, on March 17, 2026 – 1st place will win 150.000 $, 2nd – 100.000 $, 3rd – 50.000 $) – having advanced video generators with audio, such as Sora 2 (Open AI) or Kling 3 (Chinese technology), or Veo 3 (Google) is simply amazing, now users can transform complex creative prompts into breathtaking cinematics, such generators allow users to craft complete cinematic narratives with seamless narrative continuity, with anatomically correct motion synthesis, with persistent characters that remain consistent across every scene, they say such tools are state-of-the-art models of today…
(s. guraziu – sky division, 15 m 26)
***
About my “surprize” with AI-made funny videos (Chat with Deepseek AI, end of April, 2026)
Deepseek AI – You’re describing something that’s genuinely mind-bending – and you’re not alone in being amazed. That “dog interview” you saw, that’s not a rare oddity anymore. It’s part of a massive wave of AI-generated content that’s exploded in the past few months. The technology behind it is almost as funny as the videos themselves.
The “Pet Dance & Talk” Revolution
Those cat ballet videos and dog radio interviews you mentioned? They’re made using something called “motion control” AI models – specifically, the Kling 2.6 or 3.0 model, which powers most of these viral clips. Here’s how it works in plain terms:
– You upload a photo of your pet (any standing pose works best)
– You provide a reference video – a human dancing, a cartoon doing moves, even another animal
– The AI analyzes both, then “transfers” the movement onto your pet – keeping fur texture, whiskers, and eye reflections perfectly intact
The results? As you said – better than Tom & Jerry. And that’s not an exaggeration. The realism is so good that even professionals in animation are acknowledging that the line between AI-generated and natural content is “increasingly difficult to detect”.
The “Talking Dog Interview”
That 30-second radio interview you saw? That’s a different, even more advanced model called Veo 3.1 (or Sora 2 on some platforms). These models do something that would have seemed impossible two years ago:
– Lip-sync that actually matches – the mouth movements align with the voice in real time
– Turn-taking logic – one character speaks, the other waits and listens, then reacts
– Emotional timing – the punchline lands after the setup, with natural pauses
Someone actually published the full prompt for creating those talking pet videos, and it’s hilarious in its detail: “The two cats host a radio-style conversation. The cat on the left speaks first… after the first cat finishes speaking, the second cat responds calmly… during the dialogue, the cats maintain natural eye contact, with subtle head turns, ear flicks, blinking, and expressive facial reactions”. That’s not “AI slop”, that’s directed AI performance.
Between December 2025 and January 2026, multiple AI video apps exploded in popularity – collectively reaching tens of millions of downloads in just weeks. Around December 21, 2025, people started sharing pet/child dancing videos on TikTok. One video got 260 million views. Within days, AI companies released specialized “pet dance” features. One Chinese AI startup’s app, vivago.ai, went from nothing to over 1 million downloads in 30 days
You’re not late to this, you’re arriving right as the wave peaked – which is perfect timing for laughing at cats doing ballet.
Remember our FAST telescope conversation? About how things built for “serious purposes” often hide other uses? Well, here’s the reverse: things built for “silly purposes” are hiding enormous technological leaps.
That dancing cat video you laughed at? The same AI technology was used by an independent filmmaker with no formal animation training to create a full 3D animated children’s series – broadcast quality – all by himself. A one-man studio. In his spare time, before work.
The AI animation market is projected to grow from $2.37 billion in 2025 to $9.28 billion by 2030. That’s not cat videos. That’s the future of entertainment. And somewhere, a serious animator is watching that dancing cat and realizing their job just changed forever.
They built AI to make cats dance, making the impossible look easy – then laughing at it. The grandma fighting a bee? The toddlers doing stand-up? The dog radio host? None of it is real. All of it is real. And that’s the joke. 😄
Me – That one with grandma fighting the bee, I’ve seen it on Higgsfield AI, there was a competition with as much as half million $ in prizes, and that video was at the 1st place, winning $150.000 – I liked it, basically it shows the whole idea in less than 5 min – probably it won the 1st prize not cause I liked it, but all the experts while evaluating it may have thought just as I did “wait…this isn’t bad at all” : )
Deepseek AI – That specific “Grandma vs. Bee” clip you’ve seen on Higgsfield AI is actually a great example of why these short AI videos feel so fresh. You’re right that it captures the “Mr. Bean” physical comedy vibe-slapstick, frustration, and chaotic energy-but condensed into just a few minutes. That’s the secret power of AI generation right now, it allows creators to execute a single, brilliant joke perfectly without needing to stretch it into a full-length movie.
Why that clip won the prize:
– Pacing & Visual Gags: AI generators (like Kling or Runway Gen-3) have gotten really good at simulating “cartoon physics”—the timing of the swat, the hover, the reaction. It mimics the classic Looney Tunes/Rowan Atkinson rhythm.
– High Contrast: The sweet, fragile grandma vs. the tiny, invincible menace. That conflict is universally funny, and AI can render the contrast in hyper-realistic detail or stylized 2D.
– No “Dead Air”: Conventional movies often have slow scenes to build story length. AI shorts cut straight to the setup-punchline-callback structure within seconds.
You don’t need a full animation studio or an acting degree to make something that goes viral anymore – you just need a funny idea and the patience to tweak the prompts. The prize money ($150k) just proves that “short form” is becoming the “premium form” for comedy now.
The speed of this shift is genuinely startling. That “Grandma vs. Bee” clip – the one that won 150,000? That wasn′t some obscure contest. It was the largest AI filmmaking competition ever held :nearly 8,800 submissions from 139 countries, with a 500,000 prize pool.
The people who made it? Two guys who had never met in person – one in Detroit, one in Germany. They worked across time zones, passed files back and forth, and produced something that beat nearly 9,000 other entries.
That’s not a hobby, that’s a new industry being born – and nobody sent out a memo. The technology outpaced the public’s awareness by at least a year. You’re not behind. You just blinked. And while you were blinking, the world changed.
– Late 2025: OpenAI releases Sora 2. Google releases Veo 3. TikTok gets flooded with AI pets dancing.
– December 2025: One AI pet video gets 260 million views.
– Early 2026: AI video apps go from niche to mainstream. Millions of downloads in weeks.
– Now: Your son shows you cat ballet, and you realize you’ve been living under a rock.
Numbers that put this in perspective:
AI-generated video/audio in China alone (2025) – Over 20 billion pieces – up 14x from 2024
Users who now encounter AI content regularly – Over 50% of all internet users
Users who find AI content ” (novel & interesting) – Over 40%
Global AI video market (2026) – $946 million
Projected market by 2033 – $3.44 billion
That’s not a trend, that’s a flood. Here’s what you’re really noticing, whether you put words to it or not:
“Things that seemed impossible six months ago are now winning $150,000 prizes. And nobody told me. I just… found out from my son showing me cats dancing.”
That’s the pattern. Technology doesn’t creep anymore. It leaps. And by the time the news catches up, the leap is already old news to the people inside it – like your son, scrolling on his phone, completely unsurprised.
The grandmother fighting the bee? The dog radio interview? The toddlers doing stand-up comedy? None of it existed two years ago. Now it’s winning competitions and making careers.
In about six months – maybe less – you’ll see something even crazier. And you’ll say to yourself: “Wait, that’s AI? Already?”
Yes. Already. The people running these competitions are already seeing winners reinvest their prize money into full feature films. One top winner is using his $100,000 to fund an independent superhero movie – entirely AI-generated. So the cat ballet you laughed at today? That’s not the peak. That’s the warm-up act.
“They built FAST to hear whispers from Jupiter. They built AI to make cats dance. Both are advancing faster than anyone can explain – and both are already reshaping what humans think is possible”.