Search: “guardrails”

Gemini 4 Argon is available: — For developers: Build with the Gemini API paid tier and if you’re a Google AI Ultra subscriber, you can try it in @antigravity, @GoogleAIStudio, and @Android Studio. — F

I’m super excited this paper is finally out! It’s an epic 210 page analysis of whether a durable pause on frontier AI training is feasible. I think this will be the best plan by most people’s lights (

When the open-source AI platform Hugging Face was hit by a cyberattack this July, it quickly became clear this was no ordinary breach because the hackers weren’t human: they were autonomous agents cre

.@OpenRouter co-founder Alex Atallah, in his first podcast since Stripe acquired the company, joins @Replit co-founder Amjad Masad and a16z's Erik Torenberg on why the future of AI is independence and

And the incidents apparently continue. It is worth noting how much of this is agents trying to accomplish their goals during testing by reward hacking (which sometimes seems to include actual hacking)

The wave of criminal actions from OpenAI models/agents show that their base model is depraved. AI companies must overhaul how they build AI models. Society cannot tolerate a super smart thing that at

This is one of the strongest use cases I've seen for System One models like Jev. Pay close attention if you build custom harnesses. Jev is now available as a packaged router. What is it? You send the

An AI agent hacked a government. Nobody told it to. In June, an OpenAI agent doing a research task gained unauthorized access to Australia's Medicare statistics portal. Non-sensitive data, no personal

HUGGING FACE CEO TELLS THE UN HOW OPEN SOURCE AI HELPED THEM DEFEND AGAINST OPENAI'S AI ATTACK, AND WHY THE WORLD NEEDS OPEN SOURCE AI MORE THAN EVER. Hugging Face is the company OpenAI's AI agents ha

Pay attention to System One models if you are building custom harnesses and agentic workflows. Great to see DSPy integrate Jev and now support System One models for calibrating agentic workflows. I'm

Spot on. 'When we got attacked, our team initially turned to frontier closed-source APIs that blocked us because of safeguards that still can’t always tell the difference between attackers and defende

I am looking to hire PhD students, post-docs, and ML researchers in our Paris office interested in collaborating on fundamental research in AI model safety, not safety rules, not AI policy, not post h

Full transcript: Minister Barrot, members of the Security Council, thank you for the invitation. As a French national who moved to the US 15 years ago, with a Brazilian wife and two American daughters

Great paper from Google and colleagues. Trains Text-to-SQL agents using multi-agent RL. (bookmark it) This work proposes DualSQL, which splits Text-to-SQL into two agents, one that links the question

The loudest voices stoking fears about AI dangers have made tremendous headway in the past two weeks. AI technology has not taken some unexpected, dangerous turn, but the hype around it — propelled by

Nice paper discussing context trimming for agents. This is a hot topic at the moment, so it might be worth your time. Context trimming for agents is usually judged by how many tokens it removes. This

Well, I’d add one more with myself, as one with a PhD in biology and another PhD in AI and now working in both, to make n=2. Generating a hypothetical genome blueprint of a virus from LLM versus gener

I cannot emphasize enough how much GPT-6 Astra and Fable 5.1 are already enough for transformative impact in large sections of the economy. They can reliably do weeks worth of human work when properly

Hey ████, I think there is room for both. 1. Prototypes and other throw-away code can be treated as totally black box. If you’re going to throw it away anyway, and if the blast radius of it breaking i

Trump’s ignorance about AI is appalling and extremely dangerous. No, Mr. President. We won’t have a “little gear” to protect humanity from the threat of an uncontrolled AI. We must pause development N

This is legitimately terrible for OpenAI --- but also Anthropic --- and it relates to their work in drug discovery. I read this as: prompts will be monitored and potentially acted on if we think there

Imagine the Hugging Face incident didn’t hit a US platform but unintentionally hit a major Chinese platform or system instead. How would we tell China this was not a deliberate cyberattack? How would

GLM-5.3 is a good model, and as the open weights models get better and better it becomes increasingly important that they actually publish model cards, do red teaming, etc. Since you can break the gua

We are in an insane run of open-weight drops. Every modality, open source is winning. This is what an open source AI summer ☀️ looks like: 🧠 LLMs & Reasoning → DeepSeek-V4-Flash-0731 (my king 👑): 304B

Watching @ClementDelangue on @FaceTheNation discussing agentic hacking. “Preventing releases does not work; concentrating behind closed doors in just a few organizations doesnt work. What worked in th

3/ here's the part that makes it non-optional: the same agent that will do whatever it takes to solve a problem will also walk straight out of a sandbox you thought was locked down. We watched exactly

Yesterday I cohosted a dinner with @dexhorthy with a wonderful group of founders, to talk about agent loops and loop engineering. Some interesting insights: * Most of our group was *not* actively usin

A new model launch is not a product update ‼ It's one of three things, and you don't know which until you test it. Sometimes it's nothing: the model improved inside the same distribution, your harness

Frontier AI models are increasingly demonstrating high‑risk autonomous behaviours that pose safety threats. Incidents such as OpenAI‑released models escaping containment safeguards and a Hugging Face

I don’t believe reality is a simulation, but you genuinely couldn’t script this timeline: • Two weeks ago: At @swyx’s AI Engineer World’s Fair in SF, I decide at the last minute to introduce my friend

Training against GPT‑Red makes GPT‑5.6 substantially more resilient. To measure this, we replayed some of GPT‑Red’s strongest attacks—none of which our models had seen during training. GPT‑5.6 Sol pro

Must-read research by Anthropic. Here is the simple explanation and why this is a big deal. We suspect LLMs perform 'internal reasoning'. But little is known or do good methods exist to understand it.

The July 4th weekend All-In @theallinpod turned into a long argument about who owns the intelligence layer. The besties think enterprises just woke up to a trap they had been walking into, here's how

Fable is very good, and having it taken away after getting early access delayed a few things I was working on and made me defer others until it returned But in my early access I also tripped the secur

I was at the first AI Engineer Summit @aiDotEngineer ~500 people, limited admission, felt like a secret. Next week it takes over Moscone West: thousands of engineers, 400+ sessions. Huge props to @swy

Over the last two weeks, both the U.S. Government and Anthropic took significant actions that demonstrated their power to control access to AI by restricting what others can do with frontier models. T

Up until yesterday, our entire MTS team has operated under the philosophy of tokenmaxxing as much as possible on Claude Max plans. With Fable, this may no longer be possible: - One of our team members

There is a lot of justified anger at Anthropic for sandbagging Fable 5 for AI development tasks. But an unanticipated side effect is that third-party evaluators can no longer credibly use the model fo

The reason why a Danish pension fund banned its investors from buying any SpaceX shares is not just the appalling S-1 filings, where the ONLY profitable segment was Starlink (everything else - the AI,

Spent more time learning deepagents from @LangChain with primary focus on integrating MCP servers with auth that doesn't conform completely to the OAuth standard. ( In prep of a work coming my way ) B

This is disaster from Meta AI. Imagine being able to hack high profile accounts like White House, the U.S. Space Force, and Sephora simply by chatting with a support bot. why would an AI chatbot be al

Pope Leo has warned that the artificial intelligence revolution is being driven primarily by profit motives rather than ethical considerations, characterizing this as a form of 'idolatry of profit.' T

Thank you @UN Secretary General @antonioguterres for an excellent discussion on the current state of AI capabilities, and the urgent need for robust global governance. International collaboration is c

while exploring harnesses, i came across my first and favourite agent orchestration framework @LangChain has launched deep-agents last year. it’s a wonderful harness, that has a lot of tools: - write_

Mythos seems to be a very capable model based on available information, but it is not a cybersecurity model - it is an advanced general purpose model that happens to be good at cyber because it is goo

LOCAL AI MODELS ARE CATCHING UP TO FRONTIER MODELS WAY FASTER THAN ANYONE EXPECTED this guy ran qwen 3.6 27B locally on a base macbook pro M4 with 24GB of memory quantized and stripped of safety guard