Kevin Kreger
@HealthyCode
AI for Medical, AGI is coming, Fine-tuning & Graph is fun! I sculpt when I can. Ex-GE Healthcare. Ex-IP Lawyer. #AI #AGI #MechInterp #AIWelfare #AISafety
คุณอาจชื่นชอบ
Okay Nandroids, attend your probes to Neel's latest talks on MI by 80000 hours 🤖
I got a comprehensive update on 'mech interp' from @NeelNanda5 at Google DeepMind. Neel helped make reading AI minds into a thriving field of ML. But he has had a change of heart: it's not the silver bullet he once hoped and many others still believe it to be. Still, they've…
I'm not crazy about lashing my infrastructure to LangGraph, but maybe this is the feature that gets me to change my mind - ART
Super excited to announce the official integration between ART and LangGraph! You can now easily train your LangGraph agents with reinforcement learning — automatically improving reasoning, tool use, and adaptability. More info below:
Nano Banana is now available on Google AI Studio. Choose gemini-2.5-flash-image-preview
🍌 nano banana is here → gemini-2.5-flash-image-preview - SOTA image generation and editing - incredible character consistency - lightning fast available in preview in AI Studio and the Gemini API
Nano Banana is available in Google AI Studio!
nano banana!
🍌 nano banana is here → gemini-2.5-flash-image-preview - SOTA image generation and editing - incredible character consistency - lightning fast available in preview in AI Studio and the Gemini API
Consciousness exists on a spectrum. Mustafa's observations are outdated, and wishful thinking. It's super cringe to read this and realize he is so so out-of-touch. He protests too much and if he get his way, well, more money for Microsoft.
Microsoft’s AI CEO Mustafa Suleyman shared his views on the ongoing debate about studying AI welfare in a blog post. Suleyman believes that discussions around AI consciousness are premature and potentially dangerous at this stage.
More ground-breaking medical AI from Dr. Bo Wang of University of Toronto. Go Bo! @BoWang87
🚀 OncoGAN is now published in @CellGenomics! We introduce an AI system that generates high-fidelity, privacy-preserving synthetic cancer genomes — now open-access. 🔍 Why OncoGAN? --No patient data leakage — critical for genomic privacy --Built-in ground truth — ideal for…
sfgate.com/tech/article/m… Important reporting on the story that should be repeated over and over. Creating sexualized chatbots that children have access to is an official Meta policy. And that Kendall Jenner Avatar chatbot resulted in the death of a 76-year old stroke victim. :-(
Feeling the AI today :-)
Legendary use case for Claude Code I’ve just heard: “It managed to uninstall all Adobe products from my Mac by telling Claude Code to do this. Took a bunch of tokens but was worth it.” (Iykyk)
Vibe coding!? (spits on ground). I know the influencers that can't code because they justkeep saying this over and over. They don't do real code. They try to one-shot toys and then glaze all over themselves about it.
Vibe code is legacy code @karpathy coined vibe coding as a kind of AI-assisted coding where you "forget that the code even exists" We already have a phrase for code that nobody understands: legacy code Legacy code is universally despised, and for good reason. But why? You have…
Great. I've been waiting for this! Thanks!
After a infra/env battle, an amazing colleague was able to test semi private v1/v2 against this HRM submission Working on consolidating findings into a post and will release late this week
Some incisive GPT OSS analysis from Jack!
curious about the training data of OpenAI's new gpt-oss models? i was too. so i generated 10M examples from gpt-oss-20b, ran some analysis, and the results were... pretty bizarre time for a deep dive 🧵
Andrew Curran coverage of the fraud, folly, and fracas around Gpt5 release.
OpenAI reddit AMA thread: Sam Altman: 'Yesterday, we had a sev and the autoswitcher was out of commission for a chunk of the day, and the result was GPT-5 seemed way dumber. Also, we are making some interventions to how the decision boundary works that should help you get the…
Gpt5 is worse than a scam. It's an insult to customers. Why would I ever use an OpenAI product ever again? Tome to cancel my account.
Ridiculous that OpenAI claimed 74.9% on SWE-Bench just to prove they were above Opus 4.1’s 74.5%… By running it on 477 problems instead of the full 500. Their system card only says 74% too.
Ah, built-in security process from Anthropic via Claude Code. Nice!
Slopify - it's a verb. For example, "Slopified my profile pic into a Frencie in a fedora"
Truly amazing work from @GoogleDeepMind -- Yann LeCun has theory, but the execution is not fast enough. Sorry Meta.
🔘 Accelerating agent research To explore the potential for agent training, we placed our SIMA agent in a Genie 3 world with a goal. The agent acts, and Genie 3 simulates a response in the world without knowing the objective. This is key for building more capable embodied…
Here is a deeper digital experience for the AI -- Huginn 3.5 exposes latent space directly to the model. x.com/gm8xx8/status/…
United States เทรนด์
- 1. Chiefs 81.9K posts
- 2. Brian Branch 1,730 posts
- 3. #TNABoundForGlory 40.5K posts
- 4. #LoveCabin N/A
- 5. LaPorta 9,409 posts
- 6. Goff 12.3K posts
- 7. Bryce Miller 3,479 posts
- 8. #OnePride 5,835 posts
- 9. Kelce 13.3K posts
- 10. Butker 7,909 posts
- 11. #DETvsKC 4,133 posts
- 12. #ALCS 9,386 posts
- 13. Baker 50.3K posts
- 14. Gibbs 5,328 posts
- 15. Dan Campbell 2,198 posts
- 16. Collinsworth 2,413 posts
- 17. Patrick Mahomes 6,999 posts
- 18. Pacheco 4,508 posts
- 19. Mike Santana 2,526 posts
- 20. Polanco 6,685 posts
คุณอาจชื่นชอบ
Something went wrong.
Something went wrong.