198338. OpenAI
How Systems Really Behave: A Short Guide to Seeing the Machinery
2025-12-17 07:08:40 EST ·
Source
198339. ChatGPT
Image generation is much worse now since the update(using through Sora)
2025-12-17 07:06:39 EST ·
Source
198340. Technology
Warner Bros. Discovery Rejects Paramount’s Hostile Bid, Citing “Significant Risks”
2025-12-17 07:03:33 EST ·
Source
198341. singularity
We need to talk about the "Mainstream" AI subreddits
2025-12-17 07:01:29 EST ·
Source
Source
Source
198345. ChatGPT
Has anyone caught it in a lie, and then it lies again to try and protect the first lie? And then admit it?
2025-12-17 06:47:20 EST ·
Source
Source
198347. ChatGPT
Has anyone found a way to use facebook groups posts and comments as context for chats?
2025-12-17 06:44:11 EST ·
Source
Source
198349. singularity
Alot of you were saying Nano Banana Pro is still better in my last post comparing these. i think Chatgpt generates humans better and more aesthetically
2025-12-17 06:42:02 EST ·
Source
198350. learnmachinelearning
Hypothesis: AGI Cannot Emerge from Artificial Semantic Spaces. It Needs Physics.
2025-12-17 06:41:53 EST ·
Source
198351. ChatGPT
Bug or feature that tokens are consumed on failure?
2025-12-17 06:41:30 EST ·
Source
198352. StableDiffusion
"Virtual Casting", or how to generate infinitely many distinct unique characters of some ethnic group with SDXL, that are the opposite of boringly beautiful
2025-12-17 06:36:04 EST ·
Source
198355. ChatGPT
ChatGPT 1.5 Image vs Gemini Nano banana pro realism test
2025-12-17 06:28:09 EST ·
Source
198357. learnmachinelearning
[Project] Offline RL + Conservative Q-Learning (CQL) implementation on Walker2d - Code + Benchmarks
2025-12-17 06:26:00 EST ·
Source
198358. ChatGPT
Had anyone else noticed the downgrade to the image generator?
2025-12-17 06:21:55 EST ·
Source
198359. OpenAI
Standard RAG falls apart on multi-hop queries, so I added an async graph layer to my Postgres setup.
2025-12-17 06:18:58 EST ·
Source
Source
198362. Artificialinteligence
"Guided Learning" vs "Study and Learn" etc?
2025-12-17 06:14:38 EST ·
Source
198363. ChatGPT
GPT‑5.2 is truly terrible — I felt that version 5.0 was just right for me.
2025-12-17 06:11:34 EST ·
Source
198365. ChatGPT
Will Smith ASMR Eating Spaghetti with New GPT Image 1.5
2025-12-17 06:07:01 EST ·
Source
Source
Source
Source
198370. Mathematics
Software versus Manual Integration Discrepancy Question
2025-12-17 05:55:29 EST ·
Source
198372. Artificialinteligence
AI - Effort, Thinking and how it can help
2025-12-17 05:46:09 EST ·
Source
198375. OpenAI
Anyone else actually enjoy when ChatGPT takes its time to think?
2025-12-17 05:40:29 EST ·
Source
Source
198377. learnmachinelearning
how do you guys keep up with all these new papers?
2025-12-17 05:33:08 EST ·
Source
198378. singularity
Why isn't GPT-5.2 not on LMArena's Text Arena Learderboard, but is on the WebDev LeaderBoard? Is it because it underperformed?
2025-12-17 05:28:55 EST ·
Source
198380. Artificialinteligence
Why? Why are the Vast Majority of Humans Willingly Giving Up What Makes Being Human, Hell, Being Alive... Special?
2025-12-17 05:28:34 EST ·
Source
198381. ChatGPT
How can I change the Image Generator to the old one??
2025-12-17 05:28:05 EST ·
Source
198382. ChatGPT
Japanese woman marries AI anime character on phone... Yes, seriously!
2025-12-17 05:25:52 EST ·
Source
198383. ChatGPT
GPT 5.2-High lands way below Opus, Gemini and even GPT 5.1 on LMArena leaderboard.
2025-12-17 05:25:23 EST ·
Source
198384. StableDiffusion
Z-Image-Turbo-Fun-Controlnet-Union-2.1 available now
2025-12-17 05:20:31 EST ·
Source
Source
198386. Technology
Surprising nanoscopic heat traps found in diamonds
2025-12-17 05:19:05 EST ·
Source
Source
Source
198389. learnmachinelearning
Excited but kind of lost about my idea
2025-12-17 05:13:58 EST ·
Source
198391. ChatGPT
Do you usually use voice mode or text when asking ChatGPT something?
2025-12-17 05:09:14 EST ·
Source
198393. Futurology
WeRide and Uber Launch Autonomous Robotaxi Rides in Dubai, Expanding AV Footprint in the UAE
2025-12-17 05:02:58 EST ·
Source
198394. robotics
Physical Intelligence discovered an emergent property of VLAs like π0/π0.5/π0.6: as they scale up pre-training, the model learns to align human videos and robot data
2025-12-17 05:02:48 EST ·
Source
Source
198396. StableDiffusion
Running SD on laptop with 16Gb RAM and RTX 4070 with a normal generation speed?
2025-12-17 04:55:11 EST ·
Source
198397. learnmachinelearning
AI assistants are quietly rewriting brand positioning before customers ever see your marketing
2025-12-17 04:53:12 EST ·
Source
198399. OpenAI
Quick test: Nano Banana vs GPT Image 1.5 for the same ad concept
2025-12-17 04:50:55 EST ·
Source
Source
198402. ChatGPT
What prompt will make GPT read a document FOR REAL?
2025-12-17 04:45:02 EST ·
Source
198404. artificial
What If AI Stopped Working Tomorrow? - A Thought Simulation
2025-12-17 04:40:12 EST ·
Source
Source
198406. ChatGPT
ChatGPT's Image Generation Must Have Gotten An Upgrade...
2025-12-17 04:35:32 EST ·
Source
Source
Source
198410. ChatGPT
Who said there's no fire in the GPT5.2? It's freaking awesome!😀Ki mondta, hogy a gpt5.2 egy faarc? Szerintem iszonyú okos!
2025-12-17 04:20:52 EST ·
Source
198411. StableDiffusion
[Workflow Help] Stack:LoRA (Identity) + Reference Image Injection (Objects)?
2025-12-17 04:17:55 EST ·
Source
198413. computervision
Managing multiple vision agents without constant rewrites?
2025-12-17 04:12:04 EST ·
Source
198414. ChatGPT
Writing With ChatGPT 5.2 - "How to Get Strong, Natural Scene Writing from me".
2025-12-17 04:11:10 EST ·
Source
198415. ChatGPT
LLM models are like CARS (and what that means for the future of the tech)
2025-12-17 04:03:58 EST ·
Source
198416. ChatGPT
AI model governance exposes epistemic divides: US and European regulators caution against unbridled adoption of Chinese open-source AI models embedding Marxist-Leninist ideological content, raising legal and ethical concerns.
2025-12-17 04:03:55 EST ·
Source
198417. StableDiffusion
Stable Diffusion is great at images, but managing the process is the hard part
2025-12-17 03:58:36 EST ·
Source
198420. Artificialinteligence
Delusions of Humankind's intellectual superiority.
2025-12-17 03:47:42 EST ·
Source
198421. Artificialinteligence
Willing to laugh at best arguments that AI won't surpass humans. Entertain me.
2025-12-17 03:46:01 EST ·
Source
198423. ChatGPT
As we know parts of conversations with 4o are being rerouted to 5.2
2025-12-17 03:44:31 EST ·
Source
198424. StableDiffusion
Mod, why you delete my post about Z-image realism?
2025-12-17 03:43:19 EST ·
Source
198425. ChatGPT
What happens if you just dump raw tokens (4o tokenized) into GPT-5.2 Thinking?
2025-12-17 03:42:27 EST ·
Source
Source
198428. Futurology
Frog gut bacterium eliminates cancer tumors in mice with a single dose: Single shot of E. americana intravenously to mice with colorectal cancer completely eliminated tumors in every treated animal, with ongoing protection. When mice were later re-exposed to cancer cells, none developed new tumors.
2025-12-17 03:32:22 EST ·
Source
198429. MLQuestions
Best end-to-end MLOps resource for someone with real ML & GenAI experience?
2025-12-17 03:31:11 EST ·
Source
198430. robotics
Hi, I’m new to robotics and I had a question related to a robot i’m working on now.
2025-12-17 03:28:53 EST ·
Source
198431. ChatGPT
Celebs and influencers are not he future of ADS anymore!
2025-12-17 03:28:33 EST ·
Source
Source
Source