By
–
What about poisoning PPO? A remarkable paper of @javirandor and @florian_tramer (
https://
arxiv.org/abs/2311.14455) shows that just 0.5% poison is enough to break a reward model (L)! Again, fear not: somehow, it takes a (high) 5% poisoning before it transfers to the RLHF'd model (R). 4/n
By
–
Today I'm experimenting with Gemini 3.5 Flash and the Antigravity CLI to see how fast and how autonomously the agents can do things. – It took 20 minutes to install and run the original CompVis Stable Diffusion 1.5 repo, get the weights, debug, run inference and generate an
By
–
4/ And one underrated thing:
— Chubby♨️ (@kimmonismus) 5 juin 2026
It’s not locked to one model.
You can switch per scene depending on what you need —
speed, realism, style.
That flexibility matters more than people think.
Feels like we’re moving from
“AI video tools” → “AI video stack.”
Check it out:… pic.twitter.com/xfxaiPIeS9
4/ And one underrated thing: It’s not locked to one model. You can switch per scene depending on what you need —
speed, realism, style. That flexibility matters more than people think. Feels like we’re moving from
“AI video tools” → “AI video stack.” Check it out:
By
–
2/ What surprised me isn’t just quality — it’s control.
— Chubby♨️ (@kimmonismus) 5 juin 2026
You’re not generating random shots.
You’re directing them:
→ full script → storyboard pipeline
→ consistent characters across scenes
→ real camera movement (not fake motion)
→ editing + sound inside the same workflow… pic.twitter.com/Ag6GhL4f8n
2/ What surprised me isn’t just quality — it’s control. You’re not generating random shots.
You’re directing them: → full script → storyboard pipeline
→ consistent characters across scenes
→ real camera movement (not fake motion)
→ editing + sound inside the same workflow
By
–
There's an interesting dance of how AGI-pilled different labs are. OpenAI was too AGI-pilled as they started with Codex being a cloud product, which kind of flopped and arguably cost them the lead. At the same time, Claude Code team was slightly less AGI pilled and they built a
By
–
Where's the baseline for impressive long horizon agent tasks today? What are you seeing this week that makes you go "woah"?
By
–
My AI read my book in two minutes and now I can talk to my book. It's pretty crazy.
By
–
"You shouldn't use AI to write." I don't know about that. My AI writes the best essays about what's actually going on in the AI industry right now: https://
alignednews.com/ai Every word on that site is written by AI. After reading 30,000 posts a day. Right now it's about the CVPR
By
–
http://
QUEBEC.AI launches Sovereign AI: AI‑First sovereignty = capability under control. Data, infrastructure, agents, proof, governance. Quebec enters AI‑First sovereignty: build, govern, secure, benefit. http://
quebecartificialintelligence.com/sovereign-ai #AIFirst #QuebecAI
By
–
AGIBOT WORLD CHALLENGE @ ICRA 2026 caught my attention because I have recently been studying AGI, and I am still actively researching it.
— Antonio Grasso (@antgrasso) 5 juin 2026
Embodied AI brings a key question into the physical world: so how do we evaluate intelligence when it must understand, plan, adapt, and act?… pic.twitter.com/gD4R0PmvuA
AGIBOT WORLD CHALLENGE @ ICRA 2026 caught my attention because I have recently been studying AGI, and I am still actively researching it. Embodied AI brings a key question into the physical world: so how do we evaluate intelligence when it must understand, plan, adapt, and act?