I have never actually used any of them, just read about them. My understanding is that it’s still very buggy for most serious applications. Its biggest advantage, IMHO, is not being locked to any particular vendor.
INNOVATION
-
OpenAI halts Sora as Grok raises new funding round
By
–
y'a @openai qui arrête sora et grok qui remet des pièces dans la machine https://t.co/FJh0HRU14W
— ⚡Jessy SEO formation #IA générative #chatGPT 🦊 (@jessyseonoob) 24 mars 2026y'a @openai qui arrête sora et grok qui remet des pièces dans la machine
-
RISC-V: The Dark Horse of Semiconductor Racing
By
–
RISC-V is still the dark horse of the semi race.
-
SambaStack: Hardware-Software Stack for AI Inference
By
–
Simplify AI with SambaStack, the top hardware & software stack built for AI inference. Deploy on-prem or in the cloud to accelerate innovation with dedicated SambaNova infrastructure. Learn more: https://
sambanova.ai/products/samba
stack?utm_source=x&utm_medium=organic&utm_campaign=enterprise
… -
Jensen Huang to Deliver Commencement Address at Carnegie Mellon
By
–
Visionary leaders do more than build companies—they shape the foundation for the future. 🎓 Our founder and CEO, Jensen Huang, will deliver the 2026 Commencement address at @CarnegieMellon, where he will also be awarded an Honorary Doctor of Science and Technology degree. Learn more: nvda.ws/4dH4O9S [Translated from EN to English]
-
OpenAI Sora shutdown signals larger industry crisis ahead
By
–
this is like when you get a little 3.1 earthquake right before the one that gets a Wikipedia article. Culture Crave 🍿 (@CultureCrave) OpenAI is shutting down its AI video platform Sora after it flopped Disney is now exiting their $1B deal with OpenAI — https://nitter.net/CultureCrave/status/2036545881038790963#m
→ View original post on X — @timnitgebru, 2026-03-24 21:16 UTC
-
ColBERT retrieval model balances performance and computational efficiency
By
–
ColBERT seems to hit a nice sweet spot for retrieval. Easy to train and still gets great results, especially when you don't want to go full cross-encoder compute 🙂
-
Training Data vs True Intelligence: Why Real AGI Matters
By
–
You might ask, if competence can be achieved either way (by exhaustive preparation, or by having higher intelligence), why would we even care about creating actual intelligence? Isn't collecting dense enough training data good enough to achieve the goal? Intelligence is a
-
OpenAI’s Sora Team Shifts Focus to World Models Research
By
–
OpenAI's Sora team is now working on world-models – they prioritize longer-term world simulation research especially as it pertains to robotics. tl;dr what we know so far: – Sora has been cancelled because they needed the compute for their new LLM
– they renamed product -
OpenAI AGI Claims: Real Achievement or Elaborate Troll?
By
–
Either OpenAI officially achieved AGI or this is the biggest troll move ever: – they rename product organization to "AGI Deployment"
– Altman says the next LLM is a "very strong model"
– it very much accelerate the economy Quote: "Altman also said that the company would be