Thought SHAP made your fraud model explainable enough? Turns out there's a much bigger hole underneath. Title: What SHAP Can't Explain About Agentic AI Fraud URL: ❓ How does fraud detection spot "human-ness" in the first place? 💡 Typing dynamics, device switching, timing quirks: signals of human stress or deviation from a personal baseline. That assumption has underpinned fraud detection for a decade. ❓ What does SHAP actually give you? 💡 Ranked feature attribution over amount, timing, device, and velocity, explaining why a transaction looked suspicious in a way compliance analysts can audit and defend. ❓ What changes when the actor is an autonomous agent? 💡 No circadian rhythm, no panic-driven mistakes, superhuman consistency. The whole premise that "deviation from a human baseline signals fraud" simply stops applying. ❓ So what's the fix? 💡 The author argues for shifting from transaction-level scoring to "trajectory auditing", tracking an agent's decision path and tool calls, though no implemented solution exists yet. #FraudDetection# #ExplainableAI#

Kite Community Growth Plan AI Sharing ✅ Topic: AI Ethics & Human-in-the-Loop 1⃣The fairness paradox: equality of outcome vs. equality of opportunity, AI can’t solve this alone 2⃣Bias detection tools & Explainable AI (XAI) are the how 3⃣Defining fairness is the why, and that’s always a human job 4⃣Red teams, model revisioning & value alignment: the future of responsible AI Thanks @YoBroGRone for the great sharing session!👏 Better AI isn’t just better code. It’s better guardrails. #KiteAISharing# 🪁

More communications. More channels. More risk. Compliance teams need smarter ways to surface the alerts that matter most. That's why Bloomberg Vault expanded its suite of AI-powered surveillance models with two new policies for Personal Trading and Insider Dealing. Together, the suite helps financial institutions identify a broad range of potential risks across electronic communications, including market conduct, non-market conduct and conflicts of interest. Vault's AI-powered models help compliance teams: ✅ Detect potential risks across employee communications ✅ Improve alert relevance while reducing false positives ✅ Strengthen surveillance across electronic and voice communications within a single workflow ✅ Gain greater transparency with purpose-built, explainable AI models As part of Bloomberg's broader Compliance Solutions suite, Vault helps firms strengthen communications governance, streamline compliance workflows and proactively manage regulatory risk across the trading and investment lifecycle. Learn more about how Bloomberg Vault is helping firms modernize communications surveillance. 👉

Grounding an AI agent's answers in verifiable, explainable facts — an open-source platform offering the full knowledge-graph + GraphRAG + agent stack 🕸️ Title: trustgraph-ai/trustgraph URL: 🕸️ Overview An open-source semantic deployment platform for AI agents. Its core is the "context graph" — a structured, queryable representation of domain knowledge. It delivers the full agentic stack — context graphs, memory, retrieval, orchestration, and inference — for deterministic agent workloads. ❓ Challenges Solved With an LLM alone, it's hard to trace why you got an answer, and hallucination is a risk. ・Grounding an agent's answers in verifiable, explainable facts is difficult ・TrustGraph combines knowledge-graph construction with GraphRAG so agents access context that is semantically rich and verifiable ・And it runs in private deployments with sovereign control 💡 Key Features ・Multi-model DB (tabular, KV, document, graph, vectors) with multimodal support and automated entity/relationship extraction ・DocumentRAG, GraphRAG, and OntologyRAG pipelines, plus 3D GraphViz visualization ・Single/multi-agent with ReAct, Plan-then-Execute, and Supervisor patterns, and MCP integration ・Context Cores: bundle schema, graph, embeddings, evidence, and retrieval policies — versioning context like code 🌍 Tech Stack / Usage Storage on Cassandra, Qdrant, and Garage; messaging via Pulsar and others; LLMs from Anthropic/OpenAI/Google etc. plus local inference (vLLM/Ollama, etc.). Configure via npx @trustgraph/config and use the UI on port 8888. Apache 2.0 licensed. #GraphRAG# #KnowledgeGraph#

I build small, practical AI systems. On-device chatbots without LLM APIs. Embeddings, BM25, small policy networks, explainable responses. 100+ free browser-side tools for AI, math, dev, security, SEO, and weird utility work. GitHub:

Your voice agent works in the demo. The question is what happens when a real caller introduces complexities that you didn’t test for. 𝗩𝗮𝗽𝗶 𝗦𝗶𝗺𝘂𝗹𝗮𝘁𝗶𝗼𝗻𝘀 𝗶𝘀 𝗹𝗶𝘃𝗲. Build, test, iterate, and monitor in one platform: Native to Vapi, with full context of your agents: → AI testers with personalities + scenarios modeled on your real callers → tool mocks for failure paths you can't produce on demand → explainable pass/fail results with full transcripts Know your agent works before your customer finds out it doesn't. Read the blog:

📖 On-Chain Fund Source Risks Behind OKX’s Risk Control Strategy OKX’s CEO @star_okx recently stated that deposits from high-risk addresses may trigger stricter #AML# reviews, and that funds from Telegram escrow groups, Huiwang and its variants carry higher source-of-funds risk. Risk control is moving upstream — from account-level #KYC# to where the money came from, and what it passed through. Blacklist screening is still necessary. But it is no longer enough. Effective on-chain risk control needs to answer four questions: • Does the address itself pose a risk? • Is the exposure direct or indirect? • How did the risky funds flow in? • Is the risk still ongoing? That’s what @MistTrack_io built for: address screening, multi-hop exposure analysis, fund tracing and continuous monitoring — covering 19 chains, 100+ tokens, 18 stablecoins, 500M+ labeled addresses, 10K+ entities, 500K+ threat intel records and 25 risk types. In practice, MistTrack helps institutions identify source-of-funds risk before funds enter the business system: • Screen addresses by entity, behavior and threat intelligence • Measure direct vs indirect exposure with amount, proportion and hop-based analysis • Trace how risky funds entered, moved and exited • Keep monitoring after the first check These capabilities are built for high-risk deposits, P2P transfers, OTC flows and cross-chain movements — turning a risk flag into a reviewable, explainable source-of-funds path.✨ Read the full analysis👇