Season's Savings | ₹5/mo for 3 months

₹199/month after first 3 months. Renews automatically. Terms apply.
Checks & Algorithms cover art

Checks & Algorithms

Checks & Algorithms

Written by: Hosted by Anand Raghu
Listen for free

About this listen
Checks & Algorithms is a podcast about AI governance: who sets the rules for artificial intelligence, how those rules get made, and whether they actually work. Each episode unpacks the policies, regulations, standards, and corporate practices shaping AI, from the EU AI Act and U.S. policy to model audits, risk management, and accountability inside organizations. Clear, practical conversations for policymakers, practitioners, builders, and anyone curious about keeping powerful algorithms in check.Hosted by Anand Raghu
Episodes
  • Who is liable when an AI system fails - (AIGP BootCamp Chapter 7)
    Nov 11 2026

    Who is liable when an AI system fails: the developer, the deployer, or the user? In this episode of Checks & Algorithms, we break down Chapter 7 of the IAPP AIGP Body of Knowledge, Ongoing Legal, Ethical & Societal Governance Issues. Built for AIGP certification candidates and enterprise risk stewards.


    We open with a quick Chapter 6 recap: developer vs. deployer duties (EU AI Act Articles 14 and 26), the Three Lines of Defense, algorithmic impact assessments and TEVV, and post-deployment monitoring with drift detection and kill-switches.


    Then we dig into Chapter 7:

    • Liability & risk allocation: how tort and product liability law must adapt to non-deterministic AI outputs across developers, deployers, and end users
    • Web scraping rules: the CFAA and CCPA Section 7012(h) notice requirements in the US vs. GDPR Article 6 lawful bases in the EU
    • Model & data licensing: IP rights in AI outputs, performance warranties and indemnities, data protection when vendors fine-tune on your data, and training on cleared data with verifiable audit trails
    • Workforce transformation: AI literacy, upskilling vs. reskilling, and the human skills AI can't replace
    • Human alternatives: the right to opt out of automated decisions in favor of a human, per the White House Blueprint for an AI Bill of Rights
    • AI auditing: building a third-party audit profession and the accountability markers that trigger mandatory audits


    We close by tying all seven chapters together, from AI foundations and Responsible AI principles through laws, standards, the development life cycle, and enterprise risk management to long-term societal governance, wrapping up the AIGP boot camp series.

    Show More Show Less
    23 mins
  • Implementing Responsible AI Governance & Enterprise Risk - (AIGP BootCamp Chapter 6)
    Nov 4 2026

    Episode Overview

    This Audio Overview breaks down Chapter 6: Implementing AI Governance & Enterprise Risk Management from the IAPP Certified AI Governance Professional (AIGP) Body of Knowledge. Designed for certification candidates and enterprise AI stewards, the episode examines how organizations translate abstract ethical principles into operational risk controls—covering the division of governance responsibilities across developers and deployers, organizational infrastructure, testing protocols, and post-deployment deactivation procedures.

    1. Introduction & Chapter 5 Recap

    • A brief review of the four phases of the AI development life cycle: Planning (business case, scope, initial governance), Design (data governance, PETs like Federated Learning and Differential Privacy, accuracy vs. interpretability trade-offs), Development (model engineering, feature selection, training, TEVV), and Implementation (Model Cards, deployment, monitoring for failures).

    2. Chapter 6: Enterprise Risk Management & Infrastructure

    • Developer vs. Deployer vs. User: Developers design safe models, run conformity assessments, document specs, and build human oversight such as Article 14 "stop button" procedures under the EU AI Act. Deployers evaluate use-case context, manage vendor terms, assign human overseers, and execute suspension controls under Article 26. End-users rely on transparency notices, opt-out rights, and human fallback mechanisms.
    • Governance Infrastructure: Accountability across the CPO, Chief Ethics Officer, Responsible AI Office, AI Governance Committees, and Ethics Review Boards.
    • The Three Lines of Defense: First Line (business owner, holds the "Yes"), Second Line (risk, compliance, model validation, providing effective challenge and the "Not Yet"), and Third Line (internal audit, assessing gate completeness and inventory accuracy without veto power).
    • Risk Mapping & Impact Assessments: A centralized AI inventory, harm taxonomies, probability/severity matrices, a risk mitigation hierarchy, and Algorithmic Impact Assessments built on DPIAs.
    • TEVV: Red teaming, bug bashing, and threat modeling; failure mode analysis covering brittleness, hallucinations, bias, catastrophic forgetting, and false positives; Model Cards, Fact Sheets, and Counterfactual Explanations.
    • Post-Deployment Monitoring & Deactivation: Data and model drift detection, champion/challenger models, and kill switches backed by pre-committed policies and quantitative thresholds.

    3. Closing Reflection & Chapter 7 Preview

    • From Principles to Practice: Effective governance separates approval authority from stop authority, replacing "committee theater" with pre-committed stop triggers and centralized portfolio oversight.
    • Coming in Chapter 7: Legal allocation of risk and tort liability, web scraping rules under the CFAA, CCPA, and GDPR Article 6, model licensing and IP in outputs, and workforce reskilling with mandatory non-AI human fallback alternatives.
    Show More Show Less
    25 mins
  • How AI Is Built and Governed - (AIGP BootCamp Chapter 5)
    Oct 27 2026

    Episode Overview

    Good AI governance is built into the lifecycle, not added at the end. This episode of Checks & Algorithms walks through the four phases of AI development, Planning, Design, Development, and Implementation, and the governance, privacy, testing, and monitoring controls that belong in each. It is designed for candidates preparing for the IAPP AIGP certification and for leaders responsible for governing AI from idea to production.

    Topics Covered

    • Phase 1, Planning: Defining the business case, scope, and requirements; setting up governance roles, cross-functional oversight, and risk tolerance; and running an early impact assessment to confirm AI is the right solution, identify applicable laws, and map risks.
    • Phase 2, Design: Data strategy and governance, including lawful collection, quality, provenance, and lineage; data cleansing and representative labeling; privacy by design through minimization, anonymization, synthetic data, federated learning, and differential privacy; and model selection, including the trade-off between accuracy and explainability.
    • Phase 3, Development: Feature engineering that avoids spurious correlations and historical bias; model training across learning approaches; and testing, evaluation, verification, and validation (TEVV), covering performance, security, bias, fairness, and interpretability, with counterfactual explanations and documented results.
    • Phase 4, Implementation: Readiness assessments, model cards, and conformity requirements before release; deployment to cloud, on-premise, or edge environments with human oversight; monitoring for model and data drift; champion and challenger models; ongoing red teaming and security testing; and clear procedures to roll back or deactivate a model when it fails.

    Intended Audience

    • AIGP Candidates: Coverage aligned with the AIGP body of knowledge on governing AI development and the AI system lifecycle.
    • Data Science and ML Engineering Managers: Practical ways to build privacy, security, and ethics checks into MLOps pipelines.
    • Risk, Audit, and Compliance Professionals: A clear view of model card expectations, TEVV, and post-deployment monitoring.


    Checks & Algorithms is a podcast on AI governance, covering the policies, standards, and accountability practices shaping artificial intelligence.

    Show More Show Less
    21 mins
adbl_web_anon_alc_button_suppression_t1
No reviews yet