AI Deals should be explainable. “Why am I seeing this?” “Why this time window?” “What expires?” “Can I resell it?” Service Time is designing for consumer clarity from the start. #ExplainableAI #Deals

Very pleased to share that I will deliver the workshop: “AI in Biomedicine: From Data to Clinical Guidelines” at the 10th Panhellenic Conference of Molecular Medicine & Biomedical Research, Athens, 25–27 June 2026, Greece. The workshop will focus on FAIR/reproducible biomedical data analysis, explainable AI, multimodal models, computer vision, language models, and the translation of AI outputs into diagnostic/prognostic tools, clinical guidelines, and education. I would be very happy to welcome oncologists, pharma professionals, translational researchers, clinical trial teams, medical affairs colleagues, and anyone interested in clinically meaningful AI applications. Special thanks to Prof. Michalis V. Karamouzis for the kind invitation. Workshop: myrtalycongress.gr/workshop-… Conference: e-myrtaly.gr/gr/imbe2026 Website: bluesman79.github.io/ #AIinMedicine #ClinicalAI #MedicalAI #Oncology #CancerResearch #PrecisionOncology #Pharma #DrugDevelopment #ClinicalTrials #TranslationalMedicine #BiomarkerDiscovery #MedicalAffairs #MachineLearning #ExplainableAI #DigitalHealth #IMBE2026 #Athens

🔗 GraphRAG — combining LLMs with Knowledge Graphs for precise, multi-hop reasoning, explainability, and trustworthy answers in complex industrial environments. Just read this excellent technical white paper from @aasaitech on moving beyond traditional vector RAG to structured entity-relationship traversal. Key highlights: • GraphRAG vs Vector RAG: Superior precision, multi-hop reasoning, reduced hallucinations, full traceability • Core flow: Query → Entity/Graph Retrieval (multi-hop, path search) → LLM Reasoning → Answer with Evidence Paths • Industrial gold: Equipment hierarchies, failure mode analysis, RCA, compliance tracing, maintenance planning, impact analysis • Building & maintaining domain KGs from manuals, logs, CMMS, SOPs tools (Neo4j, LlamaIndex, LangChain) This is a powerful addition to the full series — elevating RAG, long-term memory, hybrid AI, agents, and safety into explainable, production-grade systems for manufacturing and edge orchestration. Full white paper infographic: x.com/aasaitech/status/20656… How are you using GraphRAG or knowledge graphs in your systems — hybrid vector graph retrieval, full multi-hop traversal for RCA, or integrated with multi-agent setups? #GraphRAG #KnowledgeGraph #RAG #IndustrialAI #ExplainableAI #AgenticAI #ManufacturingAI #EdgeAI

🔗 Hybrid AI Architectures — the mature, production-ready approach that combines the strengths of LLMs with Symbolic AI (rules & knowledge graphs), Classical ML, and Optimization Solvers for truly trustworthy industrial systems. Just read this excellent technical white paper from @aasaitech that perfectly caps the entire series. Key highlights: • Layered stack: LLM (reasoning/planning) Symbolic (rules/explainability) ML (prediction) Optimization (scheduling/solvers) • Industrial workflows: Maintenance planning, RCA, safety/compliance, supply chain optimization • Core principles: Right tool for the job, explainability by design, human-in-the-loop, guardrails, continuous feedback • Why it wins: Higher reliability, lower hallucinations, better compliance, auditable decisions This hybrid foundation elevates everything discussed in the series — MoE/RAG/CoT/agents/multimodal/edge deployment — into deployable, high-impact systems for manufacturing and edge orchestration. Full white paper infographic: x.com/aasaitech/status/20656… How are you approaching hybrid AI in your systems — heavy LLM rules/KG, full neuro-symbolic stacks, or integrated with optimization solvers? #HybridAI #NeuroSymbolic #IndustrialAI #AgenticAI #ManufacturingAI #ExplainableAI #EdgeAI

Honoured to have been invited as a Keynote Speaker at ICDAM 2026, hosted by London Metropolitan University. Today, I had the opportunity to present my keynote, "From Black Box to Glass Box: Explainable AI as the Precondition of Trust in High-Stakes Decisions," highlighting why transparency and explainability are becoming essential for trustworthy AI in medicine, agriculture, and defence. Proud to represent both Burdur Mehmet Akif Ersoy University and my postdoctoral research at the University of Nottingham on an international stage. Research has no borders, and neither should scientific collaboration. 🇹🇷 🇬🇧 #ICDAM2026 #KeynoteSpeaker #ExplainableAI #XAI #ArtificialIntelligence #TrustworthyAI #UniversityOfNottingham #Research @makuedutr @UniofNottingham

Darcy is not exactly a black box. He is a communication failure that creates black-box effects. People see his outputs, but not his reasoning. #XAI #ExplainableAI

Aporia a dual‑layer intelligence system where AI generates inferences and Aporia governs them enforcing epistemic rigor, uncertainty checks, and logically defensible reasoning. #AIResearch #ReasoningSystems #ExplainableAI #UncertaintyQuantification #AI linkedin.com/feed/update/urn…

HyperNorm AI Raises $2.2M Seed #AI #DecisionIntelligence #WealthManagement #FinancialAdvisors #ExplainableAI #InvestmentTechnology #FinancialServices #WealthTech #Fintech #Seed #Funding #HyperNormAI thesaasnews.com/news/hyperno…

Researchers, industry leaders, policymakers, innovators, and stakeholders are kindly invited to attend the Infodemic Analytics Africa (IAA) Symposium taking place at the University of Pretoria on 24 July 2026. The IAA Symposium is a research-driven workshop and collaborative platform dedicated to advancing computational methods for analysing, detecting, and mitigating narrative manipulation in African low-resource languages. IAA aims to bridge computational rigor with contextual African realities by promoting AI systems that are linguistically grounded, ethically sound, and socially responsible. 📍 University of Pretoria 📅 24 July 2026 ⏰ 09:00 – 16:00 Applications close 15th June Register here: tinyurl.com/2et6mtsc We look forward to engaging discussions, interdisciplinary collaboration, and meaningful contributions toward responsible AI and multilingual misinformation detection in Africa. #InfodemicAnalyticsAfrica #AI #DataScience #NLP #ExplainableAI #MisinformationDetection #AfricanLanguages #MachineLearning #Research #UniversityOfPretoria #GenerativeAI #LowResourceLanguages #ComputationalLinguistics #ResponsibleAI

True AI trust isn't built on faith; it's built on absolute model explainability and deterministic governance. TOPOSMIND digitalizes the thinking process, generating immutable audit trails for every choice. DM us to establish trust. #AITrust #ExplainableAI #ToposMind

🚀 Key Read in #AI Research 🩺📸 Better medical image classification—with clarity. UMAC explainable AI boosts accuracy by 1.89% across 9 datasets, while making decisions transparent. #MedicalAI #ExplainableAI #DeepLearning Read more: doi.org/10.3390/ai5040111

🇦🇺 AUSTRALIAN TECH NEWS: META ESCALATES DISPUTE AS DATA SOVEREIGNTY MOVES TO THE FOREFRONT The ongoing tension between major technology companies and the Australian government has entered a new phase. Meta has recently voiced strong opposition to Canberra's proposed News Bargaining Incentive (NBI) legislation, which would impose a 2.25% levy on domestic revenue generated by digital platforms that refuse to pay for local news content. 📢 Meta described the proposal as “grossly unfair” and argued that the measure could conflict with commitments under the Australia–United States Free Trade Agreement. ⚠️ A Warning Sign for Australian Businesses The latest dispute echoes events from 2021, when Meta temporarily blocked news content across Australia in response to earlier media bargaining regulations. From a business governance perspective, the situation highlights a growing strategic risk: platform dependency. When organizations build their distribution channels, customer engagement, and data infrastructure entirely on foreign-owned digital platforms, they are effectively operating on rented ground. Regulatory changes, commercial disputes, or geopolitical tensions can disrupt data flows and customer access with little warning. 📋 Growing Pressure from New Compliance Requirements Alongside platform-related risks, Australian businesses are also navigating an increasingly complex regulatory environment surrounding data governance and AI. 🔹 10 December 2026 Deadline Amendments to Australia's Privacy Act will require organizations to clearly disclose what personal data their AI systems use and what decisions those systems help make. 🔹 APRA CPS 230 Requirements Financial institutions must now treat third-party technology and AI services as a core operational risk, placing greater accountability on management for oversight, resilience, and risk management. 📈 Why Data Sovereignty Is Becoming a Strategic Priority Recent developments suggest that Data Sovereignty is rapidly moving from a technical consideration to a boardroom priority. Maintaining independent data infrastructure within Australia, combined with Explainable AI (XAI) frameworks and blockchain-based auditability, is increasingly being viewed as a practical approach to strengthening compliance, operational resilience, and business continuity. 🚀 Navigating Australia's increasingly complex AI, privacy, and operational risk requirements demands more than compliance, it requires the right technology strategy. From Explainable AI and blockchain audit trails to independent data infrastructure, Varmeta supports organizations in building secure, transparent, and future-ready digital ecosystems. 👉 Visit var-meta.com to explore how technology can become a competitive advantage rather than a compliance challenge. #TechNewsAustralia #DataSovereignty #AICompliance #BigTech #AustralianBusiness #TechPolicy #DataGovernance #ExplainableAI #Varmeta

An alert without evidence creates more work. Raptorx.ai transforms alerts into evidence ready intelligence with contextual risk signals, linked entities, & decision rationale available in seconds. Watch. #Raptorxai #FraudDetection #ExplainableAI #RiskIntelligence

What is Explainable AI (XAI)? Explainable AI makes AI decisions understandable to humans. Users can see why a model made a certain prediction or recommendation. #GÜRİŞTeknoloji #GÜRİŞTechnology #GTEK #guris #güriş #technology #ai #explainableai

How can pathologists gain confidence in AI-powered diagnostics?  Join Emre Köse, Computational Pathologist at Deciphex, for “Model Interpretability and Explainable AI” at the Diagnexia Computational Pathology Course. Explore: 🔬 Saliency maps 🔬 Grad-CAM 🔬 Attention visualization 🔬 Trustworthy AI systems Learn how explainable AI is helping make computational pathology more transparent and clinically reliable. Register: news.diagnexia.com/4eikfnX #ComputationalPathology #DigitalPathology #AI #ExplainableAI #Pathology

La IA explicable no debería limitarse a mostrar estadísticas o pesos neuronales. El verdadero reto es revelar la estructura de significado que conduce a una decisión. Llamo Noodion a esa unidad conceptual emergente: el puente entre cálculo y comprensión #AIEthics #ExplainableAI

Why are we testing Verbis Graph on a High-Performance Computing (HPC) environment? Many people associate supercomputers with physics simulations, climate modeling, or aerospace engineering. And they're right. But HPC is becoming increasingly important for the next generation of AI systems as well. At Prodigy AI Solutions, we chose to benchmark and evaluate Verbis Graph on HPC infrastructure because enterprise AI requires more than fast answers. It requires trustworthy answers. Testing on HPC allows us to: ✅ Evaluate retrieval performance on large-scale datasets ✅ Measure scalability across millions of relationships and documents ✅ Benchmark GraphRAG and ontology-enhanced retrieval under demanding workloads ✅ Validate multi-hop reasoning and knowledge graph traversal at scale ✅ Optimize retrieval efficiency before responses ever reach an LLM For sectors such as healthcare, life sciences, finance, legal, engineering, research, and public administration, accuracy is often more important than generation speed. Researchers have relied on HPC for decades to advance science, medicine, weather prediction, and engineering. We believe AI retrieval systems should be held to the same standard of rigorous evaluation. Our goal is simple: Build AI systems that don't just generate answers, but retrieve the right knowledge behind those answers. That's why Verbis Graph is being tested on HPC infrastructure. Because trustworthy AI starts with trustworthy retrieval. #GraphRAG #KnowledgeGraph #Ontology #EnterpriseAI #RetrievalAugmentedGeneration #HPC #Supercomputing #MachineLearning #ArtificialIntelligence #Research #DataScience #Innovation #KnowledgeManagement #ExplainableAI #AIInfrastructure #ProdigyAISolutions #VerbisGraph

What if a synthetic child patient could help us explore leukemia treatment scenarios before reaching the clinic? 🧬🤖 This is the core idea behind STING DSS. I am excited to announce that the STING Decision Support System is now live on the Web! 🚀🌐 Developed in my TÜBİTAK 1001 project, STING focuses on one of the most challenging and meaningful intersections of today’s research: Childhood acute leukemia, drug repositioning, digital twins, synthetic patients, and explainable artificial intelligence. STING DSS is designed as a browser-based AI decision-support research platform for pediatric acute lymphoblastic leukemia. It brings together multiple AI/ML modules into a sequential and interactive pipeline: 🧠 Bi-LSTM-based drug repositioning 💊 Candidate drug assessment, including Copanlisib and Novobiocin 🩸 Patient-specific pharmacokinetic/pharmacodynamic ODE simulations ⚙️ Genetic algorithm-based dose optimisation 🌐 GNN-driven digital twin prediction 🔍 Explainable AI with SHAP, permutation importance, counterfactual explanations, and GEMEX 🧬 CTGAN-based synthetic patient cohort generation 📊 5-class pediatric ALL risk stratification: LR / SR / IR / HR / VHR But beyond the technical pipeline, the real purpose is bigger: Can we make biomedical AI research more accessible, interactive, reproducible, and ethically scalable? Can synthetic patients help researchers explore treatment dynamics before moving toward more sensitive real-world settings? Can digital twins and AI-based simulations support the future of precision medicine in pediatric oncology? STING DSS is one step toward answering these questions. ✨ 🎥 In the video below, you can see the system in action. 🔗 Project Website: sting.sdu.edu.tr 🔗 GitHub Repository: github.com/tubitaksting/STIN… We are also opening a participation channel for researchers and students who would like to receive an account and contribute to the STING ecosystem. 📝 Participation Form: forms.gle/1sWVzznJ5TNwWbDq9 If you are working on AI in healthcare, leukemia research, digital twins, synthetic patient modeling, explainable AI, drug repositioning, computational biology, or clinical decision support systems, I would be very happy to connect and hear your thoughts. 🌟 Your feedback, ideas, and collaborations can help shape the next steps of STING. #STING #TUBITAK #ArtificialIntelligence #DigitalTwin #SyntheticPatients #PediatricALL #AcuteLymphoblasticLeukemia #LeukemiaResearch #ChildhoodLeukemia #DrugRepositioning #ClinicalDecisionSupport #DecisionSupportSystem #DeepLearning #MachineLearning #GraphNeuralNetworks #ExplainableAI #XAI #GEMEX #CTGAN #ODE #ComputationalBiology #BiomedicalAI #AIinHealthcare #PrecisionMedicine #PediatricOncology #HealthTech #OpenScience #Research #GitHub #SuleymanDemirelUniversitesi

🚀 Key Read in #AI Research How do you know if an anomaly explanation is actually correct? 🧠A paper introduces a ground-truth evaluation framework benchmark dataset for reliable #XAI. #AnomalyDetection #ExplainableAI #MachineLearning Read more: doi.org/10.3390/ai5040117

what if a neural network could be replaced by a few if-else statements? we did exactly that! Our research extracts decision rules from a trained neural network and converts them into an interpretable rule-based model (for classification only) The result: ⚡ 4.5× faster inference 📉 54% lower memory usage 🎯 93.5% fidelity to the original network The goal wasn’t only efficiency. It was understanding why the model makes decisions. We used image steganalysis as a case study, but the approach extends to many classification problems where explainability, trust, and edge deployment matter. Big thanks to Prof. Sabyasachee Banerjee and my co-authors for their collaboration Would you trust an AI model more if you could read its logic like code? paper: ieeexplore.ieee.org/document… #AI #MachineLearning #ExplainableAI #DeepLearning #Research #EdgeAI