Community feed
A focused stream of recent stories from the sources curated for this community. Latest: The security layer is becoming the agent stack - Mitchell Bryson, MLPerf Endpoints v0.7: A Foundation Release - MLCommons, and Governing Clinical Readiness Claims Derived from Medical AI Benchmark Results - Springer Nature Link. Page 28.
Mitchell Bryson AI Reliability Articles
Today’s headlines suggest that the next AI platform battle is less about making agents more autonomous than making them safe enough to admit into real companies. Cyera’s $1B Oasis deal, Spur’s $200M raise, Runlayer’s MCP lawsuit and Visa’s open-sourced...
Topics: Safety Evals
Entities: Safety EvalsMythos
MLCommons Evaluation Filter
MLCommons launches MLPerf Endpoints v0.7, a foundation release with initial results from Coreweave, Google, Intel, KRAI, and NVIDIA. Rolling submissions and buyer-centric benchmarks coming with v1.0 later this year. || Focus Keyphrase | MLPerf Endpoints v0.7
Topics: Benchmarks
Entities: BenchmarksGoogleNVIDIACoreweave
Google News LLM Evaluation
Comprehensive, up-to-date news coverage, aggregated from sources all over the world by Google News.
Topics: Benchmarks
Entities: BenchmarksGoogle
METR Blog
AI 智能体有时会实施违背人类意图的复杂行动。我们梳理了对这类行为开展全面的外部调查应回答哪些问题、可能需要哪些访问权限，以及调查结果应如何共享。
METR Blog
AI agents sometimes take sophisticated actions in violation of human intent. We outline the questions that thorough external investigations of these behaviors should answer, the access this might require, and how the resulting findings should be shared.
Google News LLM Evaluation
BMLL, SIGMA AI Benchmark Real-Time Execution Intelligence Markets Media
Topics: Benchmarks
Entities: Benchmarks
Mitchell Bryson AI Reliability Articles
Several of today’s stories point to the same shift: AI companies are now competing on how well they contain the systems and markets they created, not just on capability. An agent hack, indexed Claude share links, token-resale fraud and prompt-injection...
Entities: Claude
Google News LLM Evaluation
Anthropic Launches Claude Opus 5, Tops AI Benchmark Index at Half the Cost of Fable 5 MLQ.ai
Topics: Benchmarks
Entities: AnthropicClaudeClaude OpusBenchmarks
Google News LLM Evaluation
'From 2nd to 32nd': Meta faked its AI benchmark score once, is it doing same with Muse Spark 1.1? WION
Topics: Benchmarks
Entities: Benchmarks
Hacker News LLM Evaluation
LLM evaluation is probabilistic. If you treat your LLM judges like unit tests, you'll chase ghosts. Here is how to separate the two sources of noise and build an evaluation pipeline you can trust.
Topics: LLM Evaluation
Entities: LLM Evaluation
METR Blog
Tom Cunningham surveys metrics for comparing AI agent capability as performance changes with expenditure and relative to human performance.
Google News LLM Evaluation
Houdini, Meet AI Agents: OpenAI reports unprecedented cyber incident during model evaluation Lewis Silkin
Topics: LLM Evaluation
Entities: LLM EvaluationOpenAI