The Archive

Meta Muse users are adopting the app for shopping and personal productivity, the company said during Q3 earnings. The AI app now has over 10M users.

OpenAI outlines safeguard standards and usage policies for distribution partners to maintain consistent responsible access controls.

Startup Recursive has released Flint, a free, laptop-sized coding model built largely by its own AI research system, as an early test of AI that improves itself.

Anthropic just announced an AI agent competitor to Meta's Muse and OpenAI's Dots: Claude Duet. The new agent, designed for professionals and small business owners, is launching as a "research preview," according to Anthropic, and it's available as a standalone app on iOS and Android and through Claude on the web. The AI agent is built around Claude Opus 5.5. Here's Anthropic's description of what it can do: Each morning, it sends you a short brief on the most important items across your connected apps. Throughout the day, it keeps tabs on those apps and messages you only when something needs ...

Opinion: In a world of AI slop, showing the human hands making things helps.

As Amazon blocks AI agents from shopping on its site, Walmart is taking the opposite approach, introducing new technology that lets AI agents make purchases on customers' behalf, with controls over spending and transaction approvals.

Google Cloud will have GPT 6.1 Astra, Sol, and Luna in the coming weeks.

PHP 8.x security update patches two RCE vulnerabilities; disclosure process failed, exploits already circulating.

Trump’s EPA can't unilaterally dismiss SpaceXAI pollution lawsuit, court rules.

LongZero uses RL from self-generated tasks to improve LLM long-context reasoning without external training data or teacher models.

This is the kind of security issue that was found and prevented by AI, but is also a high risk because of AI.

Recommender systems allocate variable computational depth per item based on prediction need, improving efficiency over fixed budgets.

PhysiCalWorld calibrates world models to instance-specific physics properties via in-context prompting for improved prediction on novel scenes.

Quasipolynomial algorithm for computing approximate fixed points of ℓ∞ contractions; applications to solving simple stochastic games.

Self-normalized Bernstein inequalities for regenerative processes enable learning without independence assumptions.

Teacher-student analysis isolates effects of locality and weight sharing in CNNs, showing interaction with image correlation structure.

Environment Duels benchmark evaluates frontier LLMs on creating and solving interactive environments; reveals distinct authoring vs. solving abilities.

AI co-discovered A-greement, a consensus protocol for geo-replicated state machines, significantly reducing latency over state-of-the-art.

Quant-Arena benchmark evaluates LLM agents on end-to-end quantitative trading research with A-share market data and out-of-sample testing.

NPMLE achieves minimax optimal rates for univariate Gaussian mixture density estimation in Hellinger distance under sub-Gaussian assumption.

PAPAV framework unifies multimodal embodied agents across computer-use and robot-use via five capability functions: perceive, plan, act, process feedback, evolve.

Velocity Ambiguity Profile decomposes flow-matching loss into intrinsic Bayes-risk floor and data-reducible excess to diagnose optimization bottlenecks.

ASTEROID learns test-time exploration via asymmetric on-policy distillation, enabling agents to infer hidden task-relevant information from privileged experts.

BatchBench leaderboard measures AI robustness on biomedical images across unseen experimental batches using JUMP-CP dataset.

P2NPCBench pairs NP-complete and polynomial-time problems via AgentGAN adversarial rewriting to test LLM reasoning beyond recall on complexity boundaries.

Private Sparse RLHF applies differential privacy to sparse-reward RLHF with error bounds that scale logarithmically rather than polynomially with model dimension.

AtomicRL post-trains VLA models for long-horizon manipulation via atomic task structures, solving credit-assignment and subtask-imbalance bottlenecks in RL.

VX-One is a small typed-decision-head network for perimeter security triage, combining signature-engine speed with semantic model reasoning via calibrated primitives.