432 articles tagged with Safety/Alignment
📰 OpenAI’s Chief Futurist Is Leaving the CompanyJoshua Achiam spent nearly nine years at OpenAI researching AI safety and made a memorable appearance in the Musk v. Altman trial.📰 ...
As concerns rise about the potential misuse of artificial intelligence, Illinois has passed what an executive of one AI advocacy group called “the strongest AI safety law in the co...
An AI agent with a wallet is a loop that spends money. Give it a budget — "$50/day, max $5 per call"...
Picture a mineral processing plant at 3 AM. A methane sensor in the crusher hall starts climbing....
A language model that will not write a phishing email will still forward the confidential file, if a...
GodotPrompter v1.11.0 is live: a full Godot 4.7 alignment pass.90 updates landed across 26 skills: new 4.7 APIs (VirtualJoystick, AreaLight3D, joypad motion sensors) plus breaking ...
A first-of-its-kind series of bills in Illinois has been signed by Gov. JB Pritzker, aiming to improve safety for users of artificial intelligence. Rose Schmidt...
El lado del mal - Anthropic Cyber Jailbreak Bug Bounty & Severity Framework (Para Fable o "Mythos") https://www.elladodelmal.com/2026/07/anthropic-cyber-jailbreak-bug-bounty.html #...
Najnowszy raport brytyjskiego AI Safety Institute dowodzi, że tradycyjne testy AI nie mierzą szczytowych możliwości modeli, lecz jedynie ich ograniczoną wydajność przy zbyt małym b...
This new alignment series starts where the story of humanity does... in Africa! Take a trip with me that starts casually, but gets a little bumpy.nonzerosum.games/amicableape.... #...
Been thinking about how much of "AI safety" discourse focuses on hypothetical future risks while ignoring real present-day problems — biased outputs, API reliability, bad documenta...
This new alignment series starts where the story of humanity does... in Africa! Take a trip with me that starts casually, but gets a little bumpy. #GoalAlignment #AI #ArtificialInt...
Every week, AI does something stupid — and somebody calls it AGI. We're here to laugh at it. Artificial Idiots is an animated ...
Every aligned model pays the alignment tax. RLHF and DPO both trade reasoning capability for corporate agreeableness. Here is what that costs you — and why honest AI requires a dif...
A lot of AI safety trouble starts when a system treats a vibe like a fact.Tone, timing, hesitation, context, prior trust, contradiction, and uncertainty arrive as blends, not clean...
🧠LawZero proposes a framework for safety in AI predictors by emphasizing honest outputs over alignment with user preferences. The approach suggests that disinterested AI systems c...
Significant disparities exist in the diagnosis and clinical presentation of depression across different linguistic populations. Speech-based depression detection performs well mono...
Why did the US government ban Fable and Mythos, Anthropic's most powerful AI models? Let's find out! You can support me on ...
You built a system to surface information so a person could decide. Somewhere it started deciding for...
@pluralistic I did that AI alignment quiz going around.You're my patron saint apparently! 🫡Adding link https://bambamramfan.github.io/ai-compass/#AI #AIslop #enshittification
Every factory floor already has cameras. The problem is that nobody is watching them. A safety...
I dislike the #AI #llm term "jailbreak" for the same reason many folks dislike "hallucination." Like it it not, they are both "terms of art" in the space, so we must deal with them...
Cloud removal (CR) is essential for optical remote sensing, serving as a prerequisite for reliable downstream interpretation, such as semantic segmentation and change detection. Ho...
