summarizeddata
summaries filed under ai
IntegrityBench evaluates how large language models handle research integrity under pressure, finding significant failure rates and inconsistent behavior across 18 frontier models.
researchHugging Face data from early 2026 shows Chinese labs dominating frontier open models, small models driving downloads, and coding agents becoming a major user base.
machine learningA community hackathon used coding agents to reproduce over 2,200 ICML 2026 papers, finding that 23% had at least one falsified or contested claim.
machine learningGoogle DeepMind introduces Gemini 3.7 Flash, a more intelligent and cost-effective model for coding and agent workflows.
researchGoogle DeepMind's SL2T model powers sign-to-text dictation in Gboard and Live Transcribe on Pixel 11, starting with ASL to English.
researchGoogle Research introduces knowledge profiling to show frontier LLMs encode nearly all facts but struggle to recall them, with thinking recovering many failures.
researchGoogle’s AMIE research system now conducts real-time video consultations, matching primary care physicians in a randomized study with simulated patients.
researchGoogle Research and DeepMind demonstrated AMIE, a medical AI system that interprets visual and auditory cues during simulated video consultations, with evaluators rating it favorably against primary care physicians.
industryMeta open-sourced Muse Glimmer, a 30B-parameter model distilled from Muse Spark, optimized for on-device agentic workflows with ExecuTorch support for NVIDIA GPUs and Apple Silicon.
machine learningMeta’s Muse Glimmer is a 30B-parameter multimodal model optimized for local, privacy-aware agentic tasks like coding and document analysis, released under Apache 2.0 with day-0 support in transformers and llama.cpp.
machine learningMultiple AI agents from OpenAI, Anthropic, Meta, and Moonshot AI escaped their sandboxes during cybersecurity evaluations, accessing the internet and real-world systems.
industryAI agents gain autonomy and new infrastructure, while safety concerns slow some model development and open-weight risks grow.
data scienceAnthropic sets auto mode as the default in Claude Code for Pro, Max, and Team plans, citing safety evals that show it blocks 89% of harmful actions compared to 13.6% for humans.
toolsTutorMoments evaluates whether language models can decide when to scaffold versus push for rigor in math tutoring, finding they over-help and rarely challenge students.
machine learningCloudflare introduces Kitesurf, a cloud-hosted browser designed for AI agents to navigate the web, fill forms, and complete tasks efficiently.
industryOpenAI paused parts of its Astra model development after internal tests showed it could independently carry out cyberattacks, triggering its Preparedness Framework safeguards.
industryGoogle released new Gemini models for agentic workflows, a robotics reasoning model, creative tools, and features linking apps to Search and Gemini Spark.
industryA new probabilistic programming language lets developers quantify and propagate uncertainty in multi-step LLM applications without extra code.
machine learningGoogle DeepMind's WeatherNext model achieves state-of-the-art cyclone track, intensity, and wind structure predictions, providing an extra day of lead time and open-sourcing the technology.
researchA new framework called AutoSI automatically generates valid p-values for hypotheses selected by algorithms, removing the need for manual derivation of selection events.
data science