A developer typed a TV model number into Claude Code. The agent found the set on his Wi-Fi, removed 25 Xiaomi packages, and wrote a new launcher. That is…
We evaluate AI in unpredictable multi-agent simulations. Compare frontier models on collaboration, economy, conflict, logic, and creativity in live Minecraft worlds.
Vercel Board is an experimental primitive: a shared message board where agents coordinate tasks, exchange context, and work together across your tools.
Nvidia chip packaging is expanding at SPIL, where Nvidia will fund lines for bumping, interposer wafer fabrication, assembly and testing, with SPIL planning to triple advanced packaging capacity by mid-2029. Neither company disclosed the interposer type, license terms, baseline capacity or dollar
Anthropic is expected to price its blockbuster IPO at or above the top of its marketed range, after the offering drew orders for multiple times the shares available. Bloomberg’s Bailey Lipschultz and Bloomberg Intelligence’s Christopher Cain break down what could become the largest IPO fundraising ever and why history suggests getting shares at the IPO price can make a major difference to investor returns. They join Ed Ludlow on "Bloomberg Tech." -------- Like this video? Subscribe to Bloomberg Tech on YouTube: https://www.youtube.com/channel/UCrM7B7SL_g1edFOnmj-SDKg
Vision-Language-Action (VLA) models have emerged as a promising paradigm for general-purpose robotic manipulation, yet their performance often degrades on long-horizon tasks that require multiple interdependent manipulations. Instead of burdening the VLA policies with long-horizon reasoning, memory, action generation, progress tracking, and error recovery, we argue for leveraging their short-horizon manipulation capabilities within an execution-aware framework. To this end, we introduce \textnormal{\textbf{RePlan-VLA}}, a VLA-agnostic framework that translates a VLA's short-horizon capabilities into reliable long-horizon execution. RePlan-VLA explicitly monitors the robot's motion for progress evaluation, and adaptively revises the future plan based on observed outcomes. We preserve the underlying VLA policy entirely, requiring no additional fine-tuning. Experiments on RoboCasa365 and CALVIN benchmarks show an improvement of 21.9% and 10.4% in success rate over the next strongest baseline. We also validate our approach for real-world deployment on a Franka Research 3 robot.
Prompt injection is the dominant unsolved security risk for LLM agents, and the community now agrees that robustness must be measured against adaptive attackers rather than fixed benchmark injections. Many adaptive red teamers exist, and most are built to maximize attack success rate (ASR). Yet ASR counts how often a target breaks, not in how many different ways. A red teamer can achieve a high ASR by reusing a few techniques that work, leaving the others untested and absent from the attacks used to harden models. We present AutoPI, an adaptive red teamer designed to find many different ways to break a target, not just to break it often. AutoPI uses PIAtlas, a taxonomy of prompt-injection techniques, both to guide its search and to measure coverage. For each scenario, an attacker LLM writes an injection for every sub-technique in the taxonomy and then reworks the ones that fail using feedback from an LLM judge. The method is black-box and training-free, it plugs into existing benchmarks in place of their fixed attack templates, and it stays up to date as the taxonomy evolves. We evaluate AutoPI on AgentDojo and IPI Arena against open-weight and proprietary target models. Alongside ASR, we report technique coverage, the fraction of taxonomy classes, techniques, and sub-techniques that yield at least one successful injection. On AgentDojo, AutoPI finds a successful injection against Qwen3.6-35B in all six classes of the taxonomy and for more than 90% of its techniques. Against Claude Haiku 5.5, the most robust target we evaluate, it achieves between 20% and 50% higher ASR and between 25% and 400% higher class-level coverage than recent state-of-the-art adaptive red teamers, while requiring no training. We release our code and the generated attacks to support reproducible, coverage-aware evaluation.
Former Nvidia CEO Jensen Huang agreed to brief Congress behind closed doors on Dec. 1 over his affair with a suspected Chinese spy. House Democratic leader Hakeem Jeffries called it “the appropriate next step” as lawmakers demand answers about the scandal. -------- Like this video? Subscribe to Bloomberg Tech on YouTube: https://www.youtube.com/channel/UCrM7B7SL_g1edFOnmj-SDKg
The startup says its agents understand the full context of deals that bankers and consultants are working on and can automatically update financial models and presentations when new developments affect their analysis.
An open-source B-rep geometry kernel in safe Rust with zero dependencies, covering booleans, fillets, sweeps, STEP, and exact measurement. MIT licensed.
383 agent workspaces, the apps where you run, watch and steer coding agents, plus the CLI agents, editors and builders around them. Real screenshots, prices and release activity, one listing per product.
Nvidia beat expectations and raised its growth outlook, but questions remain over life after Jensen Huang. KeyBanc Capital Markets Managing Director John Vinh discusses the CEO search, talent risks and Nvidia’s push to secure more AI chip capacity with Ed Ludlow on "Bloomberg Tech." -------- Like this video? Subscribe to Bloomberg Tech on YouTube: https://www.youtube.com/channel/UCrM7B7SL_g1edFOnmj-SDKg
Existing latent reasoners typically run for a preset latent horizon during inference. Intuitively, a longer horizon should provide more opportunity for reasoning. However, in this work, we find that extending latent Chain-of-Thought (CoT) to the full preset horizon may in fact degrade reasoning performance. We observe that latent CoT evolves through distinct functional stages, with earlier computation forming frontier reasoning states, while later computation mainly reorganizes existing candidates and can weaken gold candidate support. These findings motivate a rethink of fixed-horizon latent inference. To address this, we characterize the reasoning benefits of latent CoT by realized and remaining constructive progress. Empirically, we find that higher realized constructive progress corresponds to higher current reasoning accuracy, while more remaining progress corresponds to larger future gains from continued latent CoT. To this end, we introduce a lightweight method that estimates the remaining reasoning progress and adaptively adjusts the latent horizon during inference. Across representative latent CoT reasoners and multiple arithmetic benchmarks, our method improves reasoning accuracy by up to 7.2 percentage points while reducing latent computation by up to 29.0%, without retraining the LLM.
Frontier post-training follows a common recipe: train one expert per domain with reinforcement learning (RL), then consolidate the experts into one model via multi-teacher on-policy distillation (MOPD). Such consolidation is expensive: the student samples in every environment again, an expert scores every sampled token, and the student inherits the drift of the final experts. Yet we argue that the experts' RL runs have already produced what consolidation needs: rollouts with their sampling probabilities and rewards, in the order the experts learned from them. We propose multi-expert replay policy distillation (MRPD), which replays these training logs through the student in the experts' order. At each step, the policy gradient is recomputed on these scored trajectories, and the final experts' distributions are distilled at the logged prefixes. Following the experts' own trajectory reduces the training-inference mismatch and keeps the student within the clip range, so the replay is off-policy in origin but quasi-on-policy throughout consolidation. MRPD samples nothing and calls no environment or reward model: the trainer is the sole process, and consolidating the experts costs only about a seventh of the compute used to train them. On three long-horizon agentic domains (reasoning, search and SQL), MRPD yields the best student model, substantially outperforming online RL baselines, and surpassing MOPD from the same experts at a fraction of MOPD's cost. Our analysis on the same logs shows that the replayed trajectories carry most of the gain, and that both the policy gradient and the distillation term add to it. Most interestingly, the order of the logs matters more than their amount or rewards, suggesting that expert training logs are a curriculum, not a data mixture.
As large language models (LLMs) and LLM-based agents are increasingly deployed, monitoring their behavior becomes important for detecting unsafe or unintended actions. Activation probes offer an efficient approach, but existing methods face trade-offs between accuracy, interpretability, and computational overhead. Probes on raw activations lack interpretable inputs, whereas sparse autoencoder (SAE) probes use interpretable features but often rely on additive classifiers and large SAE encoders, limiting interpretability and increasing memory overhead. We introduce GateWatch, a compact, interpretable classifier that expresses its entire decision rule as an explicit Boolean logic over a small number of binarized SAE features. We explore three approaches to constructing these circuits under a specified gate budget. For efficient deployment, GateWatchServe computes only selected SAE features, enables streaming of their binary values across layers during inference, and evaluates multiple GateWatch classifiers in parallel using compact lookup tables. Across 24 tasks on Gemma-2-9B, GateWatch achieves 80.49% mean test accuracy using only 3.8 SAE features and 3.5 logic gates per probe on average, just 1.17 percentage points below full 16K-feature SAE probes, while matching or outperforming them on 10 tasks and reducing probe memory by 99.98%. Inside vLLM, 1,000 concurrent multi-layer, multi-token GateWatch classifiers retain 97-98% of baseline throughput and 99.8% of KV-cache capacity.
Revenue of $115.8 billion, up 103% from a year ago Data Center revenue of $107.6 billion, up 110% from a year ago Vera Rubin NVL72, the most versatile and advanced computing platform eve
To get content containing either thought or leadership enter:
To get content containing both thought and leadership enter:
To get content containing the expression thought leadership enter:
You can enter several keywords and you can refine them whenever you want. Our suggestion engine uses more signals but entering a few keywords here will rapidly give you great content to curate.