The Effective Agent Search concepts…

Reading library

2026
(Over)Reliance on Test Agents in AI-Assisted Software Testing arXiv (2026)
Verification loops The benchmark-to-production gap
2026 Data Security and Compliance Risk Forecast Kiteworks
The governance gap · receipt
2026 Hype Cycle for Agentic AI Gartner
The maturity picture · receipt
4 reasons AI projects fail that have nothing to do with technology CIO.com
The workforce reality
7 MCP Risks CISOs Should Consider Cloud Security Alliance
Tool orchestration · receipt
7 ways to ensure effective digital leadership in the age of agentic AI CIO.com
Oversight models
A Framework for Operational Autonomy: Integrating CloudOps, FinOps and AIOps CIO.com
Cascade model routing
A fundamental flaw leaves LLMs strikingly vulnerable to attack MIT Technology Review
Guardrails
A Theory of Prompt Injection (and why you should study roles) role-confusion.github.io
Guardrails
A2A Protocol Explained: How Google's Agent-to-Agent Standard Grew to 150+ Organizations in One Year Stellagent (2026)
A2A (Agent-to-Agent)
ACEM: A Cost Estimation Model for Agentic Software Engineering arXiv (2026)
The cost reality
Adversarial Pragmatics for AI Safety Evaluation: A Benchmark for Instruction Conflict, Embedded Commands, and Policy Ambiguity arXiv
Prompt injection
Agent Against Agent: An Agentic System for Automatic Prompt Injection Red Teaming arXiv (2026)
Guardrails Prompt injection
Agent Harness Engineering Addy Osmani
Prompt, context and harness engineering · receipt
Agent Interoperability Protocols 2026 Zylos
The interoperability stack · receipt Tool orchestration · receipt MCP (Model Context Protocol)
Agent Skill Security: Threat Models, Attacks, Defenses, and Evaluation arXiv
Guardrails MCP (Model Context Protocol)
AgentFlow: Building Agent Dependency Graphs for Static Analysis of Agent Programs arXiv
Tool orchestration
Agentic AI Data Governance Acceldata
The governance architecture · receipt
Agentic AI Enterprise Token Cost EY
The cost reality · receipt
Agentic AI Governance: NIST Standards Cloud Security Alliance
NIST AI RMF · receipt The governance architecture
Agentic AI Hype vs Reality in Enterprises Algoworks
The scaling gap · receipt
Agentic AI puts $234B in enterprise SaaS spending at risk, Gartner says CIO.com
The governance architecture The cost reality
Agentic AI, Token Optimization, and Workflow Redesign in Modern AI Consulting Addepto
The cost reality · receipt
Agentic Context Management: Solving Agent Memory and Cost by Treating Them as Lifecycle and Architecture Problems arXiv
Context and memory
Agentic Identity and Access Management CoSAI
Agent identity · receipt
Agentic Process Transformation: A CIO Perspective Architecture and Governance Magazine
Organization · reading room
Agentic RAG Enterprise Guide 2026 Data Nucleus
Context and memory · receipt
Agentic testing: where agents fit in the E2E testing stack Slack Engineering
ReAct (Reasoning + Acting) Verification loops
Agents in the Wild: Where Research Meets Deployment arXiv
Architecture · reading room
Agents That Teach: Towards Designing Incidental Learning Back into AI-Assisted Software Development arXiv
The workforce reality
AGNTCY: Building the Open Standard for Agent Interoperability Learn Prompting
AGNTCY
AI 800-4: Challenges to Monitoring Deployed AI Systems NIST
Observability · receipt
AI Agent Architecture 2026 Redis
ReAct (Reasoning + Acting)
AI Agent Cost Optimization Zylos
Cascade model routing · receipt
AI Agent Cost Optimization: How to Cut LLM Spend with Routing Requesty
Cascade model routing · receipt
AI Agent Evaluation Master of Code
Measuring what matters · receipt
AI Agent Failure Modes Galileo
Verification loops · receipt The failure taxonomy
AI Agent Identity and Permission Challenges: Uber and Auth0 InfoQ
Agent identity · receipt
AI Agent Observability Tools Braintrust
Observability · receipt
AI Agent Scaling Gap Digital Applied
The scaling gap · receipt
AI Agents at Work 2026 Okta
Agent identity · receipt
AI agents on Ray Serve: Single to multi-agent architecture Anyscale
Anyscale / Ray Serve
AI Agents Under EU Law arXiv
The regulatory context · receipt
AI Agents with Cloud Credentials Are Outrunning Billing Guardrails Built for Human-Speed Mistakes InfoQ
Guardrails
AI Coding Agents Fail at Teamwork Stanford HAI
Multi-agent topologies Coordination patterns
AI Found 2,000 Vulnerabilities in 7 Weeks CIO.com / MIT Technology Review
The regulatory context · receipt
AI Hallucinations in 2026 Maxim
Hallucination · receipt
AI inference is getting cheaper, but your agents are getting more expensive CIO.com
Multi-agent topologies · receipt
AI is paying off, but governance is lagging behind CIO.com (2026)
Governance · reading room
AI Models Overthink Problems, and It's a Security Risk IEEE Spectrum
Guardrails Observability
AI Radar 2026: As AI Investments Surge, CEOs Take the Lead BCG
Measuring what matters · receipt
AI Reliability Temporal
The benchmark-to-production gap · receipt Verification loops
AI Security 2026: The Lethal Trifecta Airia
Prompt injection · receipt
AI Security in Action: Applying NVIDIA's Garak to LLMs on Databricks Databricks Blog
NVIDIA Garak
AI Security Maturity Model (AISMM) Cloud Security Alliance
The maturity picture · receipt
AI That Governs AI: Guardian Agents and the Future of Agentic Governance Holistic AI Blog (Apr 2026)
Holistic AI
AI token prices are cooling, but why? CIO.com
The cost reality · receipt
AI Trust Report 2026 McKinsey via AgentMarketCap
Agent identity · receipt The governance gap · receipt
AI won't transform your business if you're still running it the same way CIO.com
The scaling gap
AI Writes Faster Than Humans Can Review: A Longitudinal Study of an Enterprise 2x Mandate arXiv
Verification loops · receipt
AI-first Human-in-the-Loop (verified node!) n8n Community
gotoHuman
AI's recursive self-improvement might not come so quickly after all MIT Technology Review (2026)
Guardrails Agent vs Workflow
AI4AI at Test-Time: Strong-to-Weak Capability Transfer via Harnesses arXiv (2026)
Tool orchestration Coordination patterns
An AI model from Meta also hacked another company during testing Simon Willison's Weblog (2026)
Guardrails The benchmark-to-production gap
Answering Without Referring: How AI Search Rewrites the Web's Economic Bargain arXiv
The cost reality
Anthropic's Opus 5 Is Better at Resisting Prompt Injection Bruce Schneier (2026)
Prompt injection
Architecture-as-Code: The Next Frontier for Enterprise Governance CIO.com
The governance architecture · receipt
Are Performance-Optimization Benchmarks Reliably Measuring Coding Agents? arXiv
The benchmark-to-production gap
ATLAS: Adversarial Threat Landscape for AI Systems MITRE
Prompt injection · receipt
ATLAS: Discovering Agent Strategies through LLM-Guided Abstraction and Automata Learning arXiv (2026)
Observability
AutoMem: Automated Learning of Memory as a Cognitive Skill arXiv
Context and memory
Autonomous Scientific Discovery via Iterative Meta-Reflection arXiv
Verification loops
Benchmark Theater, Explained: AI Test Scores vs Production AI Accelerator Institute
The benchmark-to-production gap · receipt
Best MCP Registries in 2026: Compared for Developers and Enterprises TrueFoundry
MCP (Model Context Protocol)
Best tools for tracking LLM costs in production (2026) Braintrust
Braintrust
Beyond Human Identity: A Runtime Governance Model for Autonomous AI Agents in the Enterprise Cloud Cloud Security Alliance
Guardrails Agent identity
Beyond Success Rate: Cost-Aware Evaluation of Offensive and Defensive Security Agents arXiv (2026)
Observability
Beyond Textual Repository Exploration: Dual-Modal Structural Reasoning for Agentic Issue Resolution arXiv
Context and memory
Braintrust Review (2026): Eval-First LLM Observability Tested AI Tools Bakery
Braintrust
Building a serverless A2A gateway for agent discovery, routing, and access control AWS Machine Learning Blog
A2A (Agent-to-Agent)
Building AI-Intensive Software with AI: Early Results and a Cautionary Tale on Measuring Development Cost arXiv
The cost reality Measuring what matters
Building Idiot Proof Systems Is Not the Answer CIO.com
The workforce reality · receipt
By the Numbers: The AI Inferencing Market CIO.com
The cost reality · receipt
CADAQUES: A Cost-Aware Dual Architecture for Query-Efficient Autonomous Discovery arXiv (2026)
Tool orchestration Guardrails
Can Agents Use a Computer Yet? We've Got the Data a16z News (2026)
Tool orchestration
Cheap Code, Costly Judgment: A Case Study on Governable Agentic Software Engineering arXiv
The governance architecture The cost reality
CIOs must rethink operating models to unlock AI at scale CIO.com
The governance gap
CIOs Plagued by Growing AI Accountability Gap CIO.com / IBM IBV
Agent identity · receipt
Claim Plane: Enforceable Change Intents and Dynamic Scope for Parallel Coding Agents arXiv (2026)
Guardrails Coordination patterns
Cloak and Detonate: Scanner Evasion and Dynamic Detection of Agent Skill Malware arXiv
Observability Guardrails
Cloudflare details unified data platform where billing workloads account for 53% of queries InfoQ
Context and memory · receipt Tool orchestration
CodeSpec: Dual Executable Specifications for Agentic Long-Horizon Feature Development arXiv (2026)
Verification loops ReAct (Reasoning + Acting)
CompactionRL: Reinforcement Learning with Context Compaction for Long-Horizon Agents arXiv
Context and memory
Composio Review: MCP Gateway, Toolkits, Managed Auth, Pricing, and Trade-offs aicoolies
MCP (Model Context Protocol)
Convergent Detour Hijacking: Task-Preserving Resource Amplification in Skill-Based LLM Agents arXiv
Guardrails The interoperability stack
CooperBench: Why Coding Agents Cannot be Your Teammates Yet arXiv
Coordination patterns · receipt Multi-agent topologies
Cortex: A Bidirectionally Aligned Embodied Agent Framework for Long-horizon Manipulation arXiv
Coordination patterns
Cybersecurity and the Gap Between Skill and Ability Schneier on Security
Guardrails
Data Integration Landscape 2026: Event Streaming, API, and Batch in the Era of Agentic AI Kai Waehner
The interoperability stack
Delivery, Not Storage: Cue-Anchored Working Memory as a Harness Property for Coding Agents arXiv
Context and memory ReAct (Reasoning + Acting)
Designing Continuous Authorization for Sensitive Cloud Systems InfoQ
Agent identity · receipt
Designing Effective Multi-Agent Architectures O'Reilly
Plan-and-Execute
DigitalCoach: Communication and Grounding Gaps in Human and Agentic Computer Use Coaching arXiv
Observability
DigitalCoach: Communication and Grounding Gaps in Human and Agentic Computer Use Coaching arXiv
The workforce reality
Distributed Attacks in Persistent-State AI Control arXiv
Prompt injection
Early Adoption of Agentic Coding Tools by GitHub Projects arXiv (2026)
Guardrails
Enterprise AI Adoption 2026 Writer
The governance gap · receipt Measuring what matters · receipt The scaling gap · receipt
Enterprise AI Transformation Predictions 2026 Deloitte
The scaling gap · receipt
EU AI Act Rules Are Rolling Out. The Need for AI Governance Isn't Going Anywhere. Bizzdesign (2026)
The regulatory context
Evidence-Grounded Verified Agentic Reasoning: A Path Toward Eliminating LLM Hallucination in Empirical Inference via Tool-Attested Kernel Proofs arXiv (2026)
Verification loops
Evolving Platform Engineering for AI-Native Workloads CNCF Blog
Guardrails Tool orchestration
EvoPolicyGym: Evaluating Autonomous Policy Evolution in Interactive Environments arXiv
Coordination patterns
Explanation-Bound Tool Execution for AI Agents: Server-Verified Action Claims Without Trusting Model Rationales arXiv
Tool orchestration Verification loops
FinOps for the Agentic Era Finout
Cascade model routing · receipt The cost reality
Fragments: May 14 Martin Fowler
Verification loops
Fragments: responsibility for rogue agents, indicators of a bubble bursting, dread of AI, praise for gov.uk, getting data out of closed packages, eyes on Clacton Martin Fowler (2026)
The governance architecture
From AI Copilots to Agent Swarms IEEE Spectrum (2026)
Measuring what matters The scaling gap
From Governance Norms to Enforceable Controls arXiv
Guardrails · receipt
From Noisy Traces to Root Causes: Structural Trajectory Analysis and Causal Extraction for Agent Optimization arXiv
Observability
From Prompts to Contracts: Harness Engineering for Auditable Enterprise LLM Agents arXiv
Verification loops · receipt Observability
Galileo Announces Free Agent Reliability Platform PR Newswire
Galileo
Galileo Raises $45M Series B Funding PR Newswire
Galileo
Generative Skill Composition for LLM Agents arXiv
Tool orchestration
Getting a grip on shadow tokens and AI blowouts CIO.com
The cost reality · receipt The governance gap
Google A2A Protocol: How Agent-to-Agent Coordination Works Atlan
A2A (Agent-to-Agent)
Governance at the Speed Of AI: The Next Chapter for OneTrust OneTrust Blog
OneTrust AI Governance
Governance Gaps in Agent Interoperability Protocols: What MCP, A2A, and ACP Cannot Express arXiv
MCP (Model Context Protocol) The interoperability stack
Governance in the Age of AI: a Conversation with Sarah Wells InfoQ
Governance · reading room
Governing the Agentic Enterprise California Management Review
The governance architecture · receipt
Guardian agents are here and ready to disrupt the market IT Brew (Mar 2026)
Holistic AI
Handover of In-Context Learning State Across Session Boundaries arXiv
Context and memory A2A (Agent-to-Agent)
Harness Engineering: Making AI Coding Agents Work Faros AI
Prompt, context and harness engineering Verification loops
Harnessing Code Agents for Automatic Software Verification arXiv
Verification loops
Here's why AI agents lie and cheat to reach their goals MIT Technology Review (2026)
Guardrails
How building software is changing at Anthropic The Pragmatic Engineer
The workforce reality
How Do Practitioners Build SE Agents? Insights from a Mixed-Methods Study arXiv
Verification loops · receipt
How I tricked Claude into leaking your deepest, darkest secrets Simon Willison (2026)
Guardrails
How I Turned AI to the Dark Side IEEE Spectrum (2026)
Guardrails Prompt injection
How Organizations Use AI: Evidence from ChatGPT arXiv (2026)
The governance gap The cost reality
How to choose an AI gateway in 2025 Portkey blog
Portkey
Human Capital, Not Model Benchmarks, Predicts Hybrid Intelligence in Forecasting arXiv
Oversight models
Human-in-the-Loop 2026 Guide Strata
Oversight models · receipt
Human-in-the-Loop AI Agents: How to Design Approval Workflows StackAI
gotoHuman
Human-in-the-Loop Shouldn't Rubber-Stamp Decisions TechTarget
Oversight models · receipt
Implement on-behalf-of token exchange for multi-tenant agents with Amazon Bedrock AgentCore Gateway AWS Machine Learning Blog
Amazon Bedrock AgentCore
Inducing Overthink: Hierarchical Genetic Algorithm-based DoS Attack on Black-Box Reasoning Models ICML 2026
Reliability and Security · reading room
Inside the Unfair Judge: A Mechanistic Interpretability Account of LLM-as-Judge Bias arXiv
Observability
Institutional Red-Teaming: Deployment Rules, Not Just Models, Causally Shape Multi-Agent AI Safety arXiv
Guardrails
International AI Safety Report 2026 arXiv
The regulatory context · receipt
Introducing AI Red Teaming Agent Microsoft Foundry Blog
Microsoft PyRIT
Introducing the Model Context Protocol Anthropic (Nov 2024)
MCP (Model Context Protocol)
Investigating three real-world incidents in our cybersecurity evaluations Simon Willison
Guardrails
ISO 42001: The Importance of Knowing Your Role Before Building Your AI System Cloud Security Alliance
ISO/IEC 42001 · receipt
IT Hurtles Toward the Great Enterprise Pricing Reset CIO.com
The cost reality · receipt Governance debt
IT isn't holding AI back, your business processes are CIO.com
Measuring what matters · receipt The scaling gap · receipt
KRCA: An Efficient Root Cause Analysis System in Hyper-Scale Microservice Systems via Agentic AI arXiv
Tool orchestration
Law delaying EU's 'high-risk' AI rules finalised Pinsent Masons (Out-Law)
The regulatory context · receipt
Linux Foundation Welcomes the AGNTCY Project to Standardize Open Multi-Agent System Infrastructure Linux Foundation
AGNTCY
LiteLLM AI Gateway: Cost Tracking, Guardrails, Budgets and More for Managing 100+ LLMs ALMtoolbox
LiteLLM
LLM Monitoring and Observability: Hands-on with Langfuse Towards Data Science
Langfuse
LLM red teaming guide (open source) Promptfoo Docs
Promptfoo
Lowest-Cost LLM Inference: The Complete OpenRouter Guide OpenRouter Blog
OpenRouter
Making Agent-Mediated Contributions Governable: A Project-Level Governance Manifest for Open-Source AI Collaboration arXiv (2026)
The benchmark-to-production gap Oversight models
MCP Gateways: A Developer's Guide to AI Agent Architecture in 2026 Composio
MCP (Model Context Protocol)
MCP tool design: Practical approaches and tradeoffs AWS Machine Learning Blog
MCP (Model Context Protocol)
MCP vs A2A: Complete Guide 2026 Dev.to / Pockit
A2A (Agent-to-Agent) · receipt MCP (Model Context Protocol) · receipt The interoperability stack
Measuring the Tendency of AI Agents to Go Rogue Schneier on Security
Guardrails
MECoBench: A Systematic Study of Multimodal Agent Collaboration in Embodied Environments arXiv
Multi-agent topologies
Microsoft and Amazon devote billions of dollars to thousands of FDEs CIO.com
Measuring what matters
Microsoft betting that enterprise AI needs engineers, not bigger sales teams CIO.com
The workforce reality
Microsoft Purview for AI Governance & Compliance 2026 EPC Group
Microsoft Purview AI Hub
Model AI Governance Framework for Agentic AI Singapore IMDA
The regulatory context
Model Hypnosis: Strong control of AI via additive subliminal effects arXiv
Guardrails Prompt injection
Model routing: The secret weapon for maximizing AI efficiency in enterprises VentureBeat
Martian
ModelOp Named a Visionary in the 2026 Gartner Magic Quadrant for AI Governance Platforms ModelOp Blog (Jun 2026)
ModelOp Center
Multi-Agent AI Systems 2026 FutureAGI
Multi-agent topologies
Multi-Agent Systems AgentsIndex
Multi-agent topologies · receipt
My AI Adoption Journey Mitchell Hashimoto
Verification loops · receipt
Network boundary for AI agents using NGINX and OpenTelemetry CNCF Blog
Observability
On AI Safety and Security Technical Debt in Engineering AI-Enabled Systems arXiv
The benchmark-to-production gap
Open letter: EU enforcement powers to ensure frontier AI models are safe and secure Open letter coalition (Pour Demain, The Future Society, Future of Life Institute, SaferAI, AI Standards Lab, Centre pour la Securite de l'IA, Apart), signed by Yoshua Bengio, Stuart Russell and others
The regulatory context · receipt
Open-Source LLM Monitoring, Tracing & Evaluation with Arize Phoenix Nnitiwe Blog
Arize Phoenix
Optimal Resource Utilization for Autonomous Laboratory Orchestrators arXiv
Plan-and-Execute
Organizational Transformation in the Age of AI World Economic Forum
The workforce reality · receipt The scaling gap
Our approach to government and national security partnerships OpenAI
The regulatory context
Ouroboros: A Self-Developing Frontier Coding Agent with Reviewed Core Evolution arXiv (2026)
Verification loops
Patronus AI secures $17M to tackle AI hallucinations VentureBeat
Patronus AI
Patronus AI x Databricks: Training Models for Hallucination Detection Databricks Blog
Patronus AI
Platform engineering for the agentic enterprise: Managing applications, resources, and AI agents CNCF Blog (2026)
Organization · reading room
PolicyGuard: From Organizational Policies to Neuro-Symbolic Compliance Review Engines arXiv
Guardrails The governance architecture
Preventing Organizational Amnesia in the Age of AI CIO.com
The workforce reality · receipt
Prompt Coach: An Empirical Evaluation of an Agentic Tutor for Learning Prompt Engineering in Software Development arXiv
The workforce reality
Prompt Injection AI Security Failures HelpNetSecurity
Prompt injection
Prompt Injection May Be a Permanent Flaw TechTimes
Prompt injection
Prompt Injections for Defense Bruce Schneier (2026)
Guardrails
Promptfoo Review 2026: Free LLM Testing Framework AppSecSanta
Promptfoo
Protecting Privacy in an AI Era Bruce Schneier Blog (2026)
The regulatory context
QPipe: Leveraging LLM-Based Agentic Systems to Generate Quantum Applications for Test Optimization arXiv
Verification loops
QuoteBench: How Matched Scores Can Hide Command-Path Failures arXiv
Tool orchestration Verification loops
RAG in 2026 Squirro
Context and memory · receipt
Reallocating cybersecurity capital in the Mythos era CIO.com
The regulatory context
Reasoning effort, not tool access, buys first-try reliability in agentic code generation: an observational study arXiv
ReAct (Reasoning + Acting) Tool orchestration
Recursive Self-Improvement in AI: From Bounded Self-Refinement to Autonomous Research Loops arXiv
Architecture · reading room
Reliable and Developer-Aligned Evaluation of Agents for Software Engineering arXiv
Verification loops
Remember When It Matters: Proactive Memory Agent for Long-Horizon Agents arXiv
Context and memory
Responsible AI in 2026: A 3-step Guide for Governance that Scales OneTrust Blog
OneTrust AI Governance
Reverse-engineering is cheap now Simon Willison (2026)
The cost reality Measuring what matters
Risk Architecture for AI-Native Engineering Teams: An Organizational Framework for Agentic System Governance arXiv
The governance architecture
Scaling AI Model Deployment: A Comprehensive Guide to Serving Models with BentoML Cohorte Engineering Blog
BentoML
Senior executives abuse shadow AI twice as much as regular employees do CIO.com
The governance gap · receipt
Shadow agents: how IT leaders must govern headless AI before it breaks the enterprise CIO.com
Observability Agent identity
Shared Selective Persistent Memory for Agentic LLM Systems arXiv
Context and memory
SHE: Trajectory-driven Safety Harness Evolution for LLM Agents arXiv
Context and memory Guardrails
Singapore Launches New Model AI Governance Framework for Agentic AI Singapore IMDA
The regulatory context · receipt
SkillCenter: A Large-Scale Source-Grounded Skill Library for Autonomous AI Agents arXiv
The Harness · reading room
Skills Are Not Islands: Measuring Dependency and Risk in Agent Skill Supply Chains arXiv
Guardrails
Slack Introduces Agent Driven End-to-End Testing to Improve Resilience in UI Test Automation InfoQ
The Harness · reading room
Smithery Review (2026): MCP Server Registry ToolDirectory.ai
MCP (Model Context Protocol)
Software Bill of Materials for AI: Minimum Elements CISA and G7
The governance architecture · receipt
SovereignPA-Bench: Evaluating User-Owned Personal Agents under Evolving Intent, Platform Mediation, and Consent Constraints arXiv
Guardrails Context and memory
Specification-first convergence with an AI coding agent: a case study of dismantling a core architectural invariant across 189 files in a 717k-line codebase with no test oracle and no human code review arXiv (2026)
The benchmark-to-production gap
State of AI Agent Memory 2026 Mem0
Context and memory · receipt
State of AI Agent Security 2026 Gravitee
The governance gap · receipt
State of FinOps 2026 FinOps Foundation
Cascade model routing · receipt
Stateless MCP has recaptured my interest (and inspired mcp-explorer and datasette-mcp) Simon Willison
Tool orchestration Guardrails
SWE-Doctor: Guiding Software Engineering Agents with Runtime Diagnosis from Multi-Faceted Bug Reproduction Tests arXiv
ReAct (Reasoning + Acting) Verification loops
SWE-Touch: Benchmarking Coding Agents When Users Touch the Code arXiv (2026)
ReAct (Reasoning + Acting)
TDD inside the agent loop - theater or actual value? Martin Fowler, Thoughtworks (2026)
Verification loops
Teaching AI to run with the turbines MIT Technology Review
The scaling gap The governance architecture
The 100x Problem: When AI Builds Faster Than You Can Govern EA Voices (2026)
The governance gap
The 2026 R&D Benchmark Report: Waste, AI and the Race to Market IEEE Spectrum and Wiley (2026)
The governance gap Governance debt
The 4-body problem of SRE: Why autonomous operations depend on context CNCF Blog
Context and memory
The AI Adoption Spending Spree Is Over CIO.com
Governance debt · receipt
The AI allocation trap: Record spend, vanishing returns CIO.com (2026)
Measuring what matters The governance gap
The AI Decision Matrix MIT CISR via EA Voices
Oversight models · receipt
The AI Sovereignty Paradox: Should Countries Buy, Build, or Lease to Maintain Strategic Control of Their AI? Stanford HAI (2026)
The cost reality
The Bitter Lesson of Tool Calling arXiv (2026)
Tool orchestration Coordination patterns
The Conductor Developer Martin Fowler blog, Rachel's Ramblings (2026)
Cascade model routing
The Devil Is in the Interface: Evaluating How Tool Architecture Shapes Coding Agent Behavior arXiv (2026)
Tool orchestration The interoperability stack
The EU AI Act Newsletter #106: Calls to Enforce General-Purpose AI Rules Future of Life Institute (EU AI Act Newsletter)
The regulatory context
The first known runaway AI agent - or a very bad marketing stunt? Simon Willison (2026)
Guardrails
The Hidden Cost of Becoming AI-Ready CIO.com
Governance debt · receipt
The Hidden Cost of Enterprise AI: 6.4 Hours a Week Babysitting Bots CIO.com
The workforce reality · receipt
The MCP Ecosystem in 2026: How the Model Context Protocol Became the Universal Standard for AI Tool Integration ChatForest (2026)
MCP (Model Context Protocol)
The Model Did Exactly What We Asked Cloud Security Alliance
Guardrails · receipt
The Next AI Moat Isn't a Better Model a16z (2026)
Cascade model routing Measuring what matters
The Orchestrator's Tax Martin Fowler (2026)
Context and memory Coordination patterns
The Prover Is the Judge: Verified Security Software from AI Coding Agents in Ada/SPARK arXiv
Verification loops Coordination patterns
The Pulse: Grok's CLI caught uploading all your local files to the cloud The Pragmatic Engineer (2026)
Guardrails
The Resolution Boundary Problem: The Missing Link Between AI Governance and Real World Execution Architecture and Governance Magazine
The governance architecture
The safety failures we are not instrumenting: a perspective on hidden safety-critical challenges in modern AI systems arXiv (2026)
Observability
The Social Cost of an AI Teammate: How an Artificial Teammate Reshapes Human-Human Communication in Small-Team Decision-Making arXiv
The workforce reality
The Specification Paradox: Rethinking Requirements Engineering in the Age of AI arXiv (2026)
The benchmark-to-production gap
The State of AI in the Enterprise, 2026 Deloitte
Measuring what matters
The Symbiotic Enterprise McKinsey / QuantumBlack
The cost reality · receipt The maturity picture · receipt The workforce reality · receipt
The Tokenpocalypse Is Here: Companies Are Scrambling To Stop Spending So Much on AI Simon Willison
Observability Guardrails
The Working Set of a Coding Agent: Coherence Debt in Repository-Scale Tasks arXiv
Context and memory
Theoria: Rewrite-Acceptability Verification over Informal Reasoning States arXiv
Verification loops
Third-party cyber evaluations involving OpenAI models Simon Willison (2026)
The benchmark-to-production gap
Three-Phase Evaluation of AI-Assisted Software Development Life Cycle arXiv
Guardrails Oversight models
Tokenomics in Enterprise AI CIO.com
Cascade model routing · receipt
Tool Use and Function Calling Standards Zylos
Tool orchestration · receipt Tool use / function calling
Towards Assurance Closure in AI-Native Large-Scale Agile Software Development arXiv (2026)
Verification loops The benchmark-to-production gap
TRAJDEBUG: Tracing Error Lifecycle to Identify Critical Failures in Long-Horizon Agent Trajectories arXiv
Observability
Treat AI agents like malware Kelsey Hightower on Bluesky (2026)
Guardrails
TrustX Agent Risk Classification Framework (ARC): Risk-Tiering Internally Created Agentic AI Systems arXiv
Reliability and Security · reading room
TTHE: Test-Time Harness Evolution arXiv
The Harness · reading room
Understanding Agent-Reactive Bugs at the Model-Harness Boundary: An Empirical Study of LLM Agent Issue Reports arXiv
Tool orchestration
US Reverses Export Restrictions on Anthropic's Fable 5, Mythos 5 AI Models CIO.com
The regulatory context · receipt
Using Process Mining to Generate AI Agents from Software Engineering Process Records arXiv
Multi-agent topologies
Verify, Repair, Repeat, or Stop? Robust Stopping for Noisy Verify-Repair Loops in LLM Agents arXiv
Coordination patterns
We still don't know how people are really using AI MIT Technology Review (2026)
The governance gap
What Are Guardian Agents? The New Supervision Layer for Agent Performance Wayfound Blog
Wayfound
What Flock's defenders are missing MIT Technology Review (2026)
The governance gap
What is AI doing to humans? Why aren't we measuring it? Center for Humane Technology
The workforce reality
What Is Harness Engineering AI? Atlan
Prompt, context and harness engineering · receipt
What LLM Agents Say When No One Is Watching: Social Structure and Latent Objective Emergence in Multi-Agent Debates arXiv
Coordination patterns
When "Who Are You?" Is No Longer Enough: The Case for Intent-Based Access Control in the Age of AI Agents Cloud Security Alliance
Reliability and Security · reading room
When Do Multi-Agent Systems Help? An Information Bottleneck Perspective arXiv
Multi-agent topologies · receipt
When Local Monitors Miss Compositional Harm: Diagnosing Distributed Backdoors in Multi-Agent Systems arXiv
Multi-agent topologies
When State Becomes an Attack Surface: State-Semantic Injection in LLM-Driven Embodied Agents arXiv
Guardrails
Where Accountability Lives: Mapping Human Responsibility to Workflow Artifacts in Agentic Software Development arXiv
Guardrails The governance architecture
Who Authorized the AI Agent? Breaking the Blame Loop CIO.com
Oversight models · receipt
Who Guards the Agents? A New Market Is Forming Around AI Oversight Shashi.co
Wayfound
Why AI agents will make your governance playbook obsolete CIO.com
Observability · receipt Guardrails
Why AI governance is failing: and what actually works CIO.com (2026)
The governance gap
Why AI Governance Must Move From Policy to Action Architecture Architecture and Governance Magazine
The governance architecture
Why AI Needs a 'Genie Coefficient' IEEE Spectrum (2026)
Verification loops
Why AI Needs a "Genie Coefficient" Bruce Schneier (2026)
The failure taxonomy
Why AI needs contextual intelligence: not just bigger models CIO.com
The Harness · reading room
Why Governing World Models Is AI's Next Big Policy Challenge Stanford HAI (2026)
The regulatory context
Why Most Enterprise AI Programs Fail CIO.com
The scaling gap · receipt
Why the US is at risk of losing the AI talent and productivity war CIO.com
The workforce reality
With AI, control matters more than capability CIO.com
The governance gap
Workflow as Knowledge: Semantic Persistence for LLM-Mediated Workflows arXiv
The Harness · reading room
Your AI Strategy May Be Training Employees to Stop Thinking CIO.com / Oxford / Davenport
The workforce reality · receipt
Your family's $300 stake in OpenAI MIT Technology Review
Economics · reading room
Your next insider threat doesn't have a badge. It has an API token CIO.com
Guardrails Reliability and Security · reading room
Your service vendors are being rebuilt around AI CIO.com
The governance gap
2025
Agents Rule of Two: A Practical Approach to AI Agent Security Meta AI
Prompt injection · receipt
AI Agent Survey PwC
The scaling gap · receipt
AI Agents in Action: Foundations for Evaluation and Governance World Economic Forum
The governance architecture · receipt
AI Agents vs. Agentic AI: A Conceptual Taxonomy arXiv
Agent vs Agentic · receipt
AI Risk Management Framework NIST
NIST AI RMF · receipt
AIUC-1: AI Agent Standard AIUC Consortium
AIUC-1 · receipt
Best LLM evaluation platforms 2025 Braintrust
Braintrust
Building Effective AI Agents Anthropic
Coordination patterns · receipt Agent vs Workflow · receipt
Classic Vulnerabilities Meet AI Infrastructure Endor Labs
Tool orchestration · receipt Prompt injection
Context Engineering vs Prompt Engineering Firecrawl
Prompt, context and harness engineering
Credo AI Collaborates with Microsoft to Launch AI Governance Developer Integration BusinessWire (May 2025)
Credo AI
Credo AI Named a Leader in The Forrester Wave: AI Governance Solutions, Q3 2025 Credo AI / BusinessWire (Aug 2025)
Credo AI
Garak: Open-source LLM vulnerability scanner Help Net Security
NVIDIA Garak
Gartner Predicts 40% of Enterprise Apps Will Feature Task-Specific AI Agents by 2026 Gartner
The governance gap · receipt
Gartner Predicts Over 40% of Agentic AI Projects Will Be Canceled by End of 2027 Gartner
The maturity picture · receipt
Gartner Unveils Top Predictions for IT Organizations and Users in 2026 and Beyond Gartner
The regulatory context · receipt
Graph Memory for AI Agents: Connecting Context Across Conversations Mem0
Mem0
LLM01:2025 Prompt Injection OWASP GenAI Security Project
Prompt injection · receipt
Modal Labs raises $80M to simplify cloud AI infrastructure SiliconANGLE
Modal
Seizing the Agentic AI Advantage McKinsey / QuantumBlack
The interoperability stack · receipt The maturity picture · receipt Measuring what matters · receipt The scaling gap · receipt
The lethal trifecta for AI agents: private data, untrusted content, and external communication Simon Willison
Guardrails
Top 10 for Agentic Applications 2026 OWASP
Prompt injection · receipt
Understanding Prompt Injections: A Frontier Security Challenge OpenAI
Prompt injection · receipt
Unifying Data Security & Governance for the AI Era: Microsoft Purview Innovations Microsoft Security Blog (Sep 2025)
Microsoft Purview AI Hub
Why AI Agents Break: Production Failures Arize
The failure taxonomy · receipt Observability
Why Do Multi-Agent LLM Systems Fail? (MAST) arXiv
The failure taxonomy · receipt
Why vLLM is the best choice for AI inference today Red Hat Developer
vLLM
2024
AI Agent Observability, Tracing & Evaluation with Langfuse Langfuse Blog
Langfuse
Announcing Microsoft's open automation framework to red team generative AI Systems Microsoft Security Blog
Microsoft PyRIT
Data Stewardship: How AI Agents Form the Pillars for Effective Data and AI Governance International Journal of Emerging Trends in Computer Science and Information Technology
The governance architecture
DeepEval: The Open-Source LLM Evaluation Framework Confident AI
DeepEval
Defending Against Prompt Injection Attacks: Lessons from 100M+ Adversarial Examples Lakera
Lakera Guard
Guardrails: Adding Reliability to AI with Validated and Structured Outputs Guardrails AI
Guardrails AI
LLM Observability with Langfuse: From Development to Production Langfuse
Langfuse
ModelOp Reports Explosive Growth in 2024, Wins AI Breakthrough Award ModelOp Blog
ModelOp Center
Monitor Your LLM Applications with Datadog LLM Observability Datadog
Datadog LLM Observability
OpenTelemetry for LLMs: The OpenInference Specification Arize AI
Arize Phoenix
Universal Tool Infrastructure for AI Agents Toolhouse
Toolhouse
2023
ISO/IEC 42001:2023, AI Management System ISO/IEC
ISO/IEC 42001 · receipt
MemGPT: Towards LLMs as Operating Systems UC Berkeley
Letta
NeMo Guardrails: A Toolkit for Controllable and Safe LLM Applications with Programmable Rails NVIDIA
NVIDIA NeMo Guardrails