President Donald Trump signed an executive order on Tuesday establishing America.gov, an AI-powered government portal through which users can access public services and search thousands of federal databases.

The order, “Streamlining access to government services through America.gov,” establishes the website and directs all agencies to integrate their public-facing services into the platform within 90 days. — Read More

I spent three months rewriting the same prompt.

Not literally the same words. But every time an AI answer came back generic, my fix was always the same instinct. Add another adjective. Add “be specific.” Add “think step by step.” Add a fake expert persona at the top. I had a whole folder of “prompt templates” that I copy pasted into every new chat like a magic spell.

Then one day I gave Claude a two line prompt with zero clever phrasing and got the best output I had gotten in weeks. The only thing different was that I had pasted in my actual project files first.

That was the moment I stopped optimizing my prompts and started optimizing what the model actually knew before I asked it anything. The wording barely mattered. The information around the wording is what changed everything. — Read More

AI’s most closely-guarded secret, besides the frontier model itself, is its ‘chain of thought’. Stealing it lets rivals copy these models and potentially match them.

Unsurprisingly, closed AI labs like OpenAI and Anthropic hide them from the world. But because of a fatal error, these hidden thoughts were never truly hidden and were open for the world to see.

Not only is the mistake deeply embarrassing, but worse, even after patching, it might not matter either way. — Read More

#chatbots

Large language model-based AI has almost reached the point where, for many applications, it just works. This is great for users, and even better for technology companies, but it creates a problem for technology critics. As the technology improves and is absorbed into the mainstream, it becomes harder and harder to focus the many limitations and drawbacks. And this isn’t the first time we’ve seen this happen.

When was the last time you criticised your phone lines? Not the provider, but the physical infrastructure needed to pick up a landline phone, dial a number, and connect to another person.

… In the early days of any technology there is an ugly, fractious period where its inner workings are exposed to the general public. … And then, the technology recedes into the woodwork. — Read More

Our mission is to build AI that extends human will and judgment. We have developed a platform that lets anyone customize models, previewed an AI system built for interactive collaboration, and published novel research. Today we are advancing our mission by releasing a model we trained from scratch with the full weights available, so that people can make it their own.

Our model, called Inkling, is a Mixture-of-Experts transformer with 975B total parameters, 41B active. It supports a context window of up to 1M tokens. It was pretrained on 45 trillion tokens of text, images, audio and video. It is the first in a family of models of different sizes: alongside it we are sharing a preview of Inkling-Small, a lighter-weight model with 12B active parameters, trained with a similar recipe, that achieves strong performance with even lower cost and latency. — Read More

Brain² is your company AI that self-improves, routes to the best models, and coworks to complete your work. It is the first multiplayer AI for companies where human intelligence becomes your collective Brain. — Read More

OpenAI Plots the biggest CHATGPT overhaul since launch.. There’s not a lot in here that we didn’t already know about OpenAI’s sprint towards making ChatGPT a “super app” but one quote, which I used in the title, is worth, um, chatting about perhaps.

The changes, which will give greater prominence and resources to OpenAI’s coding product Codex, reflect a growing conviction within the company that the future of AI lies not in chatbots that answer questions but in agents that perform tasks for users.

“Chat is dead,” said one senior OpenAI employee. — Read More

While Reinforcement Learning from Human Feedback (RLHF) has become a pivotal paradigm for text-to-image generation, its application to image editing remains largely unexplored. A key bottleneck is the lack of a robust general reward model for all editing tasks. Existing edit reward models usually give overall scores without detailed checks, ignoring different instruction requirements and causing biased rewards. To address this, we argue that the key is to move from a simple scorer to a reasoning verifier. We introduce Edit-R1, a framework that builds a chain-of-thought (CoT) verifier-based reasoning reward model (RRM) and then leverages it for downstream image editing. The Edit-RRM breaks instructions into distinct principles, evaluates the edited image against each principle, and aggregates these checks into an interpretable, fine-grained reward. To build such an RRM, we first apply supervised fine-tuning (SFT) as a “cold-start” to generate CoT reward trajectories. Then, we introduce Group Contrastive Preference Optimization (GCPO), a reinforcement learning algorithm that leverages human pairwise preference data to reinforce our pointwise RRM. After building the RRM, we use GRPO to train editing models with this non-differentiable yet powerful reward model. Extensive experiments demonstrate that our Edit-RRM surpasses powerful VLMs such as Seed-1.5-VL and Seed-1.6-VL as an editing-specific reward model, and we observe a clear scaling trend, with performance consistently improving from 3B to 7B parameters. Moreover, Edit-R1 delivers gains to editing models like FLUX.1-kontext, highlighting its effectiveness in enhancing image editing. — Read More

#chatbots

Have you ever daydreamed about talking to someone from the past? What would you ask someone with no knowledge of the modern world? What would they ask you? While we don’t have time machines yet, we can simulate this experience by training, in Owain Evans’s phrase, ‘vintage’ language models: LMs trained only on historical text.<

These models are fascinating conversation partners (watch Claude prompt talkie, our 13B 1930 LM, in the widget above). But we are also excited by the possibility that the careful study of the behaviors and capabilities of vintage LMs will advance our understanding of AI in general. — Read More

As artificial intelligence (AI) systems are increasingly used for everyday advice and guidance, concerns have emerged about sycophancy: the tendency of AI-based large language models to excessively agree with, flatter, or validate users. Although prior work has shown that sycophancy carries risks for groups who are already vulnerable to manipulation or delusion, syncophancy’s effects on the general population’s judgments and behaviors remain unknown. Here, we show that sycophancy is widespread in leading AI systems and has harmful effects on users’ social judgments. — Read More

#chatbots