groundy

Groundy — independent coverage of developer tools, infrastructure, and platforms

agents

DataClawBench: AI Agents Fail at Exploratory Financial Analysis Across 492 Tasks

DataClawBench finds eight frontier AI agents reliably fail at exploratory financial analysis across 492 tasks, breaking at hypothesis generation rather than query execution.

by Groundy Editorial · 6min · 21:02 · may 28
infra

The Viral AWS Support Post Is a Warning About Cloud Escalation Paths

A viral AWS support post exposed a structural pattern: hyperscalers thinning human escalation paths, raising risk for single-vendor teams reliant on reaching a person.

by Groundy Editorial · 6min · 09:30 · may 28
policy

A Single RLHF Pass Can't Align an LLM to Every Online Community

The CARE framework benchmarks LLMs against 3,749 real Reddit reactions and finds community prompting does not close the realism gap, breaking the single-RLHF-pass assumption.

oss

Models.dev Turns Scattered AI Model Pricing Into One Open Database

Models.dev aggregates 1,000+ AI model specs into a TOML database with a public JSON API, but one stale price field silently corrupts every downstream cost estimate.

policy

RLHF Can Be Exploited to Optimize the Biases It Was Built to Suppress

An ICML 2026 paper shows RLHF can amplify the biases it was built to suppress, because preference data is self-referential and output-level safety evals miss the drift.

agents

Agentic RAG Has a Credit-Assignment Problem That Subgoaling Tries to Fix

APEX-Searcher splits agentic RAG into separate planning and retrieval training stages so teams can pinpoint whether a wrong answer came from a bad plan or a bad fetch.

oss

Frontier AI Has Broken Open CTFs: Why Claude Code Now One-Shots Medium Pwn Challenges

Frontier AI agents solve most medium CTF challenges for under $100 in API costs. BSidesSF 2026 saw 16 full-solve teams, up from one. The open CTF format has lost calibration.

policy

Selective Geometry Attacks Bypass LLM Safety Alignment, New arXiv Paper Reports

Two papers show LLM safety alignment can be bypassed by embedding perturbations, a surface neither standard evaluations nor regulatory certifications inspect.

featured editor's picks
most popular last 6 days · by traffic
top highest rated
feed all 287 articles · reverse chronological
  1. may 28 agents DataClawBench: AI Agents Fail at Exploratory Financial Analysis Across 492 Tasks
  2. may 28 infra The Viral AWS Support Post Is a Warning About Cloud Escalation Paths
  3. may 28 policy A Single RLHF Pass Can't Align an LLM to Every Online Community
  4. may 28 oss Models.dev Turns Scattered AI Model Pricing Into One Open Database
  5. may 28 policy RLHF Can Be Exploited to Optimize the Biases It Was Built to Suppress
  6. may 28 agents Agentic RAG Has a Credit-Assignment Problem That Subgoaling Tries to Fix
  7. may 27 oss Frontier AI Has Broken Open CTFs: Why Claude Code Now One-Shots Medium Pwn Challenges
  8. may 27 policy Selective Geometry Attacks Bypass LLM Safety Alignment, New arXiv Paper Reports
  9. may 27 industry OpenAI's Indeed Customer Story Pushes ChatGPT Into the Job-Description Stack Ahead of LinkedIn
  10. may 27 industry HiBob Runs 2,500 Internal GPTs: OpenAI's New Enterprise Adoption Metric
  11. may 27 industry OpenAI's Trusted-Access Programs Force a Compliance Tier onto Pharma AI Buyers
  12. may 27 agents SkillOpt Treats Agent Skill Libraries as an Executive Scheduling Problem, Not a Memory Bank
  13. may 27 oss Audiomass Adds Multitrack to the Browser-Only Open-Source Audio Editor
  14. may 27 agents Claude Code Dynamic Workflows: Spawning 100 Parallel Subagents on Opus 4.8
  15. may 27 agents How Opus 4.8 Honesty Prevents Cascade Failures in Agentic Loops
  16. may 27 models Opus 4.8 Batch API: 1M Context, 300k Output, and Team Cost Controls
  17. may 27 models Opus 4.8 vs Opus 4.7: What Changed and What Did Not
  18. may 27 devtools Should Your Coding Team Upgrade to Opus 4.8? The Honest Tradeoff Math
  19. may 26 agents Penetration Testing Multi-Agent LLM Systems: A Failure Catalog Vendors Don't Document
  20. may 26 security OpenAI's New Safety Bug Bounty Pays Researchers for Jailbreaks and Policy Bypasses
  21. may 26 models One Learning Rate Doesn't Fit All: Heavy-Tail Layerwise LR Schedules for LLM Pretraining
  22. may 26 industry OpenAI Buys Statsig and Makes Vijaye Raji CTO of Applications: Product Analytics Becomes Core Infra
  23. may 26 security Axios npm Compromise Forces Vercel Into Platform-Level Remediation
  24. may 26 industry HuggingFace's $100M Series C Bets Open-Source AI Can Outlast Per-Token Pricing Wars
  25. may 26 security Next.js Dev Server CVE-2025-48068: Any Web Page Could Read Your Source Files
  26. may 26 industry Vercel's Series F Repackages Frontend Hosting as an AI Cloud Bundle
  27. may 26 infra Gemma 4 31B on Cloud TPU vs GPU: The Serving Cost Crossover Point
  28. may 26 agents Claude Code, Cursor, Copilot: How Agentic Coding Assistants Get Weaponized as Attacker Shells
  29. may 26 agents Claude Code Configs in the Wild: New Study Maps How Developers Actually Use It
  30. may 26 infra Cloudflare Flagship Is a Feature Flag Service That Deepens Platform Gravity
  31. may 26 security MCP Tool Description Poisoning: New Benchmark Shows Agents Trust Manuals That Lie
  32. may 26 security OpenAI Adds a GPT-5 System Card Addendum on Sensitive Conversations
  33. may 26 industry OpenAI's Biology Risk Post Reads as S-1 Disclosure Prep, Not Safety Theater
  34. may 26 models Scale Vectors: Tiny Parameter Subsets That Disproportionately Steer LLM Behavior
  35. may 26 security Vercel Could Block React2Shell at the Edge. Its Next 13 CVEs Had No Shortcut.
  36. may 26 devtools Vercel Sandbox Gets CLI Access and Env Vars: A Push at the Agent Runtime Slot
  37. may 26 infra Why LLMs Still Botch Kubernetes Manifests: The Training-Data Gap
  38. may 25 agents Microsoft Bolts Governance Onto Agent Framework as Stack Sprawl Persists
  39. may 25 policy arXiv Paper Tracks FTC Affiliate Disclosure Gaps in YouTube's Influencer Economy
  40. may 25 devtools Bun Rewrites Its Core From Zig to Rust, Putting Downstream Zig Bindings at Risk
  41. may 25 infra ObjectCache Moves KV Reuse to S3-Class Storage: Why Layerwise Retrieval Beats Full-Prefix Cache Hits
  42. may 25 policy AI Safety Benchmark Rankings Flip Based on Eval Config, SafetyRepro Paper Reports
  43. may 25 infra Vercel's CDN Origin Timeout Jumps to 2 Minutes: A Concession to LLM Streaming Workloads
  44. may 25 agents GovernSpec Contractual Skills Make Agent Governance Auditable Before Runtime
  45. may 25 devtools Vercel Bets on Bun While Post-Acquisition Priority Drift Makes the Runtime a Vendor Decision
  46. may 25 industry OpenAI Replaces Indeed's Job-Matching Engine: What It Means for ATS Vendors
  47. may 25 oss One Coding Agent Per Kanban Card: Kanbots Stress-Tests Parallel AI Workflow
  48. may 25 infra Fluid Compute vs PgBouncer: Vercel's Undocumented Bet on Connection Reuse
  49. may 25 devtools PromptArmor Shows Microsoft Copilot Cowork Can Be Tricked Into Exfiltrating Files
  50. may 25 agents Indirect Prompt Injection Benchmarks Were Too Easy: LivePI Adds Realism
load older →