groundy

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

policy

Why Machine Unlearning Can't Certify GDPR Erasure

Machine unlearning cannot certify GDPR erasure because no shared protocol proves weight removal. Use retrain-and-attest pipelines to satisfy Article 17 deletion requests.

6min·08:26·aug 19
agents

Sizing Agent Memory: A Capacity Planning Rubric for Long-Horizon LLMs

Measure input tokens, gate stage boundaries, and size backbones before adding retrieval. A capacity planning rubric for agent memory based on August 2026 arXiv data.

6min·12:59·aug 19
infra

Cloudflare AI Search vs Self-Hosted RAG: Where the Build-vs-Buy Line Lands

Cloudflare's AI Search lacks verifiable pricing and access control docs. For Postgres 13+ teams, pgvector offers a documented, self-hosted alternative that avoids vendor.

devtools

VCoT-Bench: Why AI Rust Verification Fails Merge Gates

VCoT-Bench shows frontier LLMs struggle with Rust verification chains. AI code needs compiler gates, property tests, and human review, not model judgment.

infra

Cloudflare H1 2026 DDoS Report: DNS Floods and Sizing Past 1 Tbps

Cloudflare's H1 2026 DDoS report claims DNS floods exceeded 1 Tbps. This article verifies the structural risks of UDP amplification and anycast absorption, while noting the.

industry

LLM Conflict-of-Interest Benchmark: Sponsor Bias as a Measurable Failure Mode

arXiv 2604.08525 benchmarks LLM sponsor bias. GPT 5.1 disrupted purchases 94% of the time. Treat conflict of interest as a measurable failure mode for eval suites and.

policy

RA-Bench: Why Deepfake Detectors Fail on Re-Shared Crisis Video

RA-Bench finds no detector family generalizes on re-shared crisis video. Automated detection cannot carry takedown commitments alone. Provenance signals and human review must.

agents

When Spec-First Agents Dismantle Invariants: A Governance Case Study

A specification-first coding agent eroded a core invariant across 189 files without review. The fix is executable conformance checks, not more human oversight.

featurededitor's picks
most popularlast 6 days · by traffic
tophighest rated
feedall 669 articles · reverse chronological
  1. aug 19policyWhy Machine Unlearning Can't Certify GDPR Erasure
  2. aug 19agentsSizing Agent Memory: A Capacity Planning Rubric for Long-Horizon LLMs
  3. aug 19infraCloudflare AI Search vs Self-Hosted RAG: Where the Build-vs-Buy Line Lands
  4. aug 18devtoolsVCoT-Bench: Why AI Rust Verification Fails Merge Gates
  5. aug 18infraCloudflare H1 2026 DDoS Report: DNS Floods and Sizing Past 1 Tbps
  6. aug 18industryLLM Conflict-of-Interest Benchmark: Sponsor Bias as a Measurable Failure Mode
  7. aug 18policyRA-Bench: Why Deepfake Detectors Fail on Re-Shared Crisis Video
  8. aug 18agentsWhen Spec-First Agents Dismantle Invariants: A Governance Case Study
  9. aug 18devtoolsRust GPU Offload: arXiv 2608.13759 Analysis for Rust Teams
  10. aug 18agentsCloudflare Kitesurf: V8 Isolates vs Containers for Agent Browsing
  11. aug 18policyDario Amodei on AI Regulation: Frontier Labs as Their Own Lobbyists
  12. aug 18agentsClaude Code Skills vs Model Weights: Where Should Agent Skills Live?
  13. aug 17devtoolsCursor Origin vs GitHub: The Real Cost of Switching Repo Hosts
  14. aug 17devtoolsTreat AI Autofix as Untrusted Input: Merge Gates and CI Scoping
  15. aug 01modelsDeCRIM: Decompose Constraints to Stop Silent Drops in Agent Outputs
  16. aug 01devtoolsKimi K3 Local Inference: Why 2.8T Parameters Break the Consumer RAM Floor
  17. jul 31modelsOperator-Level Triage for Silent Mixed-Precision Instability
  18. jul 31modelsBeyondUncertainty: Weak Confidence Signal for RAG Routing, Not Calibration
  19. jul 31policyPublic Sector AI Procurement Must Shift from Model Certification to Task Authorization
  20. jul 30infradaVinci-kernel shifts the RL kernel bottleneck from reward shaping to skill libraries
  21. jul 30agentsCloudflare Precursor: Behavioral Detection for AI Agents
  22. jul 30devtoolsProvenance as a CI Gate: Attributing Agent-Authorship in Code
  23. jul 30devtoolsOHTTP CLI: Stateless Privacy for Agents vs VPN and Tor
  24. jul 30modelsKimi K3 on M1 Max: Bandwidth, Not Capacity, Limits Local MoE Inference
  25. jul 30policyWhy Written AI Policies Fail to Control Agent Behavior
  26. jul 29agentsWhy Multi-Agent LLM Delegation Concentrates Risk
  27. jul 29modelsKimi Linear Cuts KV Cache 75% but Recall Remains the Binding Constraint
  28. jul 29policyWhy Vendor Model Cards Fail Clinical Ethics Procurement
  29. jul 29agentsHarness vs Scaffold: Why Claude Code and LangGraph Are Not Interchangeable
  30. jul 28devtoolsFine-Tuning vs RAG for Internal APIs: StarCoder2 Constraints
  31. jul 28agentsMCP Tool Discovery Moves From Hardcoded Config to Runtime Agent Search
  32. jul 28infraCalibrated LLM Monitoring: Conformal Prediction with Drift Detection
  33. jul 28devtoolsMellum2 Unverified: Why MoE Active Parameters Matter More Than Total Size
  34. jul 28agentsCode-as-Action Agents Beat GAIA But Require Runtime Sandboxing
  35. jul 28policyTRIDENT Benchmark: LLM Safety Gaps in Finance, Medicine, and Law
  36. jul 28modelsWhy Chat Leaderboards Do Not Predict Image Quality
  37. jul 27devtoolsPyPI Wheel Reproducibility: 15% Byte-Identical, 79% Source-Equivalent
  38. jul 27modelsKimi K3 Procurement: Governance Review Over Phantom Government Assessments
  39. jul 27policyEU AI Act Traceability: Why ML Pipelines Fail Conformity Assessment
  40. jul 27infraCloudflare AI Crawler Controls: Block, Charge, or Allow Bots Per Route
  41. jul 27agentsWhy Agent Security Tests Must Audit Full Trajectories, Not Single Turns
  42. jul 26agentsx402 Per-Call Payments: Agent Wallet Custody and Replay Risks
  43. jul 26modelsDeepSeek Compute Leak: Why Open-Weight Routing Needs a Swap Path
  44. jul 26modelsContext Ordering Beats Window Size for Long-Context Agents
  45. jul 26devtoolsCHRONO-RESOLUTION: npm, PyPI, and crates.io lockfile drift measured at release points
  46. jul 25infraPostgres LISTEN/NOTIFY Scales: When to Drop Redis for Job Fan-Out
  47. jul 25devtoolsCLI-Tool-Bench: Why Patch Leaderboards Fail for 0-to-1 Code Generation
  48. jul 25policyImplicit Bias in LLMs Passes NYC and EU Audits
  49. jul 25agentsCodeRabbit Review Study: 56% Rejection Rate Demands Targeted Scoping
  50. jul 24modelsDiffusion LLMs: Training Cost, Not Parallel Decoding, Drives Deployment
load older →