This is truly the video AI revolution!
GENERATIVE AI
-
Small Models AI Discussion and Use Cases Exploration
By
–
Lots of interest around small models! It was a pleasure ranting about AI and talking about your use cases. Thanks to @josephinePqt1 and the AI Circle for inviting me!
-
AI Video Generation Enters New Dimension with Higgsfield x Kling 3.0
By
–
Ce n’est pas une simple évolution.
— Jouhatsu | AI Influence Operator (@Jouhatsu_ai) 5 février 2026
C’est un changement de règle.
Higgsfield × Kling 3.0 fait entrer la vidéo IA dans une autre dimension.
⁰Séquences multi-plans. Audio natif.
Cohérence totale des personnages.⁰15 secondes de génération continue sans hacks, sans rafistolage.… https://t.co/Gge6ievNUh pic.twitter.com/hGWXPGigW2Ce n’est pas une simple évolution. C’est un changement de règle. Higgsfield × Kling 3.0 fait entrer la vidéo IA dans une autre dimension.
Séquences multi-plans. Audio natif. Cohérence totale des personnages. 15 secondes de génération continue sans hacks, sans rafistolage. -
RLHF Bias: Millions for Nice but Useless Answers
By
–
They are working on it; the problem is that they paid millions of dollars, notably OpenAI, to get RLHF databases where labelers tended to choose nice answers rather than useful ones. Whether it's human bias or a directive, we don't know, but it's been identified….
-
Goodbye verifiable Google, hello divergent AIs
By
–
Yes indeed! The growing modularity of AI models in relation to user contexts will eventually create more and more differing views of reality. Gone is the verifiable "it's true, I saw it on Google," replaced by "but my AI didn't tell me the same as yours!"
-
New Gemini model checkpoint spotted in A/B testing
By
–
BREAKING : A new Gemini checkpoint has been spotted in A/B testing. Will we see this live?
-
Advancements in AI Video Generation Quality
By
–
Holy shit the new Kling AI model is looking ultra crispy.
— Bilawal Sidhu (@bilawalsidhu) 5 février 2026
Blows my mind how quickly AI video has gone from incoherent slop to shattering the visual turing test. pic.twitter.com/l6n1MvSikuHoly shit the new Kling AI model is looking ultra crispy. Blows my mind how quickly AI video has gone from incoherent slop to shattering the visual turing test.
-
xAI Grok Imagine Generates Artistic Images and Videos
By
–
Grok Imagine is now available! xAI’s Grok Imagine generates artistic images and cinematic short videos with audio support from text, image, or video. It is built for fast creative iteration, short‑form content, multiple variations per prompt, flexible aspect ratios, and
-
Real-time world model generation from text and image prompts
By
–
In our research lab, we are building “real-time dreaming” – the ability to generate fully playable video worlds prompted from any text or image.
— Roblox (@Roblox) 5 février 2026
Our real-time, action conditioned world model (currently running internally at 16fps at 832x480p) is trained on a combination of data,… pic.twitter.com/R5CKJkFuh2In our research lab, we are building “real-time dreaming” – the ability to generate fully playable video worlds prompted from any text or image. Our real-time, action conditioned world model (currently running internally at 16fps at 832x480p) is trained on a combination of data, including proprietary Roblox 3D avatar/world interaction data. World models are different from multiplayer engines in that they store state and memory in video latents. Roblox is multiplayer, and we are actively researching optimal ways to simultaneously store state for thousands of players, and keep them in sync with their environment. Our world model leverages database technology which stores all user interactions on Roblox in a vector format that can be used to re-render video and interaction from any camera angle. We see several immediate uses for our Roblox world model. We will use it side-by-side text, image and video prompts as a way to launch auto-generation of immersive worlds. In Roblox Studio, a creator could walk around and use prompts to “paint” a world and then convert it into a 3D representation or direct to Roblox native as a way for many people to play simultaneously. All of this comes alive as we explore the notion of a “Dream Theater” – where one user is dreaming, while others watch and prompt them. 2/4 Community note: The beginning of the video has stolen assets/design from Clair Obscur: Expedition 33 by Sandfall Interactive. The environment is the Flying Waters area and the girl in the video is Maelle. store.steampowered.com/app/1903340/Cl… expedition33.wiki.fextralife.com/Flying+Waters clair-obscur.fandom.com/wiki/Maelle
→ View original post on X — @shiqi_yang_147, 2026-02-05 01:28 UTC
-
Meta’s upcoming Avocado model reported as most capable internally
By
–
According to The Information, upcoming Avocado model from Meta is referenced as the “most capable “ to date internally. Soon?