Resource Library

Explore the crucial role of AI Observability and Security in building responsible AI for trustworthy, secure, and compliant LLM and ML deployments.

Enhance AI with the Fiddler Trust Service — cost-effective LLM monitoring, the fastest AI guardrails, and VPC or air-gapped deployments for trusted AI security.

Discover how Fiddler's AI Observability and Security platform provides guardrails, monitoring, and insights to enhance predictive models, LLMs, and GenAI in production.

See how Fiddler brings enterprise-grade monitoring directly into your secure SageMaker environment.

Protect LLM apps with industry's fastest guardrails. Catch hallucinations, safety violations & prompt injections. Try Fiddler Guardrails free.

Which guardrails solution is right for your organization? One size never fits all — and the stakes couldn't be higher. Our interactive 2025 Benchmarks Report provides objective data to navigate tradeoffs and select the right guardrails for your use case.

Experience Fiddler Guardrails: Industry's fastest guardrails with <100ms response times. Secure against hallucinations, toxic content & prompt injections. Try free for 14 days.

Discover Fiddler Guardrails integration with NVIDIA NeMo, delivering industry-leading <100ms response time. See how this fastest, most cost-effective solution protects LLMs against hallucinations, safety violations, and jailbreak attempts while deploying securely in your environment.

Explore Fiddler’s AI Observability and guardrails for government agencies, ensuring AI performance, transparency, and security in mission-critical applications.

Discover how Fiddler and Domino partner to deliver trusted, scalable, and secure MLOps solutions, empowering government agencies and enterprises to operationalize AI with confidence.

Fiddler Guardrails, an extension of Fiddler Trust Service, enables real-time scoring to detect risky LLM issues like hallucinations and prompt injection, rejecting malicious inputs before they reach the model.

In this product tour, we show how to leverage existing model data to create segments based on protected attributes like gender and race, and define intersectional segments for in-depth fairness analysis.

In this product tour, see how to configure Fiddler’s dashboards for specific teams, business goals, and use cases.

In this product tour, see how to configure Fiddler’s dashboards for specific teams, business goals, and use cases.

In this product tour, we walk through tracking drift for models like image classifiers by leveraging image embeddings and text embeddings.

In this product tour, see how to build custom dashboards, visualize model performance, detect prediction drift, and dive into root cause analysis for issues like data integrity problems or feature drift.

In this product tour, see how Fiddler supports a range of ML models, including binary classification, regression, and ranking.

The Fiddler AI Observability platform helps financial service institutions monitor, analyze , explain, and improve predictive and generative AI models and applications.

Discover how Fiddler AI Observability’s advanced image explainability helps you gain deep insights into the predictions and behaviors of computer vision models.

See Fiddler in action and discover how Fiddler Trust Models detect and help you safeguard your LLM application while driving business growth.

The Fiddler AI Observability and Security platform helps insurance companies monitor, analyze, and protect predictive models, generative AI applications, and agents.

Fiddler partners with federal agencies to ensure the safety and performance of AI models, enabling responsible AI with comprehensive observability and monitoring.

Fiddler provides computer vision capabilities to aid government agencies to detect and analyze issues in image models, enhancing high-stakes decision-making.

Explore a method for measuring distributional shifts in text data using language model-based embeddings, highlighting the effectiveness of LLMs in capturing semantic relationships for this purpose.

Practical lessons and experiences on developing AI applications and chatbots using retrieval-augmented generation (RAG) methods.

In this workshop, see how Fiddler users can use the Fiddler Report Generator (FRoG) to create customizable reports for model risk management (MRM) and periodic model health reviews.

Watch this demo to explore how the Fiddler AI Observability Platform for LLMOps streamlines end-to-end LLMOps by enabling data scientists and AI practitioners.

Watch this demo to explore how the Fiddler AI Observability Platform for MLOps enables data scientists and AI practitioners to monitor, explain, analyze, and improve ML models.

Watch this panel of industry leaders share their insights on LLM implementation and their significant impact on business ROI in their respective fields.

In this workshop, Murtuza Shergadwala shares his experiences in building the RAG-based Fiddler Chatbot, and provides tips and tricks on prompt engineering, and more.

Watch this session as Josh Rubin explores the question of whether LLMs can be explained, emphasizing that understanding a model's behavior is crucial for identifying weaknesses, ensuring robustness, and aiding human decision-making.

In this fireside chat with CEOs Krishna Gade and Lukas Biewald, they discuss AI trends, leveraging LLMs, and the complexities of LLM deployment at scale.

In this session, Stanford Professor James Zou discusses how ChatGPT's behavior is changing over time. Discover the fascinating research findings and implications for AI language models.

Watch this workshop to learn the importance of evaluating LLMs using Fiddler Auditor, strategies to mitigate prompt injection attacks, and developments in LLMs.

Watch this AI expert panel discuss the advancements and challenges in LLMs from proof-of-concept to production, the complexities of LLMOps, and regulatory compliance.

Watch this expert panel discuss generative AI in enterprise, focusing on model efficiency, personalization, RAG's role, and aligning AI with business needs.

Watch this keynote from Hanlin Tang on the discussion of the science and engineering behind LLMs in enterprise applications, the choice between buying LLM services from external providers and building custom models, specialized domain-specific models, and more.

Improve LLMs with robustness testing before deploying them to production.

Learn how Fiddler’s unique, clustering-based method accurately monitors data drift in NLP models and LLM-based embeddings.

Learn more about Fiddler Auditor, the open source robustness library that facilitates red teaming of LLMs. Improve LLMs with robustness testing before deploying them to production.

Learn about Fiddler Auditor, the open source robustness library that facilitates red teaming of LLMs. Robustness testing is a critical step in pre-production to minimize hallucinations, bias, and adversarial attacks.

Learn how to track the performance of LLM-based embeddings from OpenAI, Cohere, Anthropic, and other LLMs by monitoring drift using Fiddler.

Learn how Fiddler AI Observability provides ML teams with a unified platform to deliver and improve high-performing ML models while saving time and money.

Watch this session from Ali Arsanjani, PhD, Head of the AI Center of Excellence at Google, to learn about the importance of explainability, adaptability, and risk-minimizing in enterprise generative AI.

Hear from Amit Prakash, CTO and Co-founder at ThoughtSpot; Diego Oppenheimer, Partner and CEO in Residence at Factory; and Roie Schwaber-Cohen, Staff Developer Advocate at Pinecone on how LLMOps optimizes MLOps for LLMs, the key pieces of a generative AI workflow, and how ML teams can leverage LLMs in their applications.

Saad Ansari, Director of AI at Jasper AI, dives into how AI can be used for societal benefit, the importance of intentionality when designing new technology, and applications for generative AI.

Watch this panel session with Ricardo Baeza-Yates, Director of Research at Institute for Experiential AI, Northeastern University; Miriam Vogel, President and CEO at EqualAI, Chair at National AI Advisory Committee; Toni Morgan, Responsible Innovation Manager at TikTok, to learn why responsible AI principles and frameworks are necessary, why model governance is important, and the limitations of LLMs.

Watch this panel session with Payel Das, Principal Research Staff Member and Manager, Trusted AI a IBM Research; Srinath Sridhar, CEO and Co-founder at Regie.ai; Casey Corvino, CTO and Co-founder at Lavender AI to learn how engineering teams are using generative AI to improve productivity and how IBM is using generative AI to drive medical advances.

Watch this session with George Mathew, Managing Director at Insight Partners to learn about the current state of AI, the different modalities of human and machine interaction, the advancements, and challenges AI presents.

Learn how Fiddler AI Observability provides ML teams with a unified platform to monitor, analyze, explain, and improve ML models at scale, and build trust into AI.

Learn how Fiddler’s custom dashboards help ML teams obtain actionable model insights, and increase organizational alignment and collaboration.

Learn how to quickly detect model performance and drift issues, and reduce the time to troubleshoot issues with root cause analysis using Fiddler.

Learn how to get a deep understanding of model predictions using Fiddler's explainability capabilities.

Learn how to monitor models with unstructured data using Fiddler's cluster-based binning approach.

Learn how to evaluate and validate models in Fiddler before deploying them into production.

Learn how to perform root cause analysis to gain contextual insights, and improve model outcomes using Fiddler's analytics capabilities.

Learn how Fiddler’s powerful and customizable alerts provide early warning on model performance.

Learn the human-centric challenges and requirements for ML model monitoring in real-world applications.

Learn about FairCanary, a novel approach to generating feature-level bias explanations that is an order of magnitude faster than previous methods.

Scott Zoldi, Chief Analytics Officer at FICO, has authored over 100 patents in ML and AI. He discusses why AI needs to grow up fast and what orgs can do about it.

Anjana Susarla, who holds the Omura-Saxena Professorship in Responsible AI at Michigan State, discusses the different dimensions of responsible AI.

Anand Rao, Global AI Lead, PwC discusses how organizations can implement AI responsibly.

Fiona McEvoy, founder of YouTheData.com, shares her perspective on algorithmic bias, deep fakes, emotional AI, and the way that AI systems impact our behavior.

Lofred Madzou, AI Lead at the World Economic Forum, shares his experiences managing AI projects around the world and the top things to keep in mind when implementing AI.

Listen to this episode of the Responsible AI podcast to hear how different companies are defining responsible AI and measuring the impact of AI applications.

Maria Axente, Responsible AI lead for PwC UK, shares what RAI means, why its importance is overlooked, and ways to incentivize teams to implement AI ethically.

Read about GroupShapley and GroupIG (Integrated Gradients), axiomatically justified methods to understand drift in ML model prediction distributions over time.

Improve your MLOps workflow on SageMaker using Fiddler's Model Performance Management platform.

Learn why fairness metrics need to incorporate intersectionality, including a simple method to expand the definitions of existing group fairness.

Learn how organizations view and use explainability for stakeholder consumption, including a framework for establishing clear goals for explainability.

Learn how subtle differences in the underlying game formulations of existing methods can cause large differences in the attributions for a prediction.

Learn how randomized feature ablation helps measure the importance of a feature on a model's ability to make good predictions.

An overview of model interpretability and explainability in AI, key regulations/laws, and techniques/tools for providing explainability as part of AI/ML systems