Techsplainers by IBM

This episode of Techsplainers explores AI agent swarms, groups of AI agents that coordinate their work toward a shared goal. The episode explains how swarms divide larger tasks among specialized agents, allowing different agents to plan, research, analyze, review and synthesize work. It also distinguishes agent swarms from the broader category of multi-agent systems, showing how swarms place particular emphasis on collaboration, specialization and coordinated activity between relatively autonomous agents.
The discussion examines how agent swarms divide tasks, share information and coordinate work through centralized or decentralized approaches. Listeners learn about sequential, parallel and iterative execution models, as well as collaboration patterns such as handoffs, supervisor-worker structures, debate, voting and shared-state workflows. The episode also considers the potential benefits of swarms, including speed, specialization, flexibility and scale, alongside tradeoffs such as coordination overhead, conflicting outputs, context management, higher costs and error propagation. Finally, it explains when a swarm may provide real value and when a single agent or conventional workflow may be the simpler option.
Find more information at https://www.ibm.com/think/topics/agent-swarm
Find all the Techsplainers episodes here: https://www.ibm.biz/techsplainers-podcast
Narrated by Derek Robertson

This episode of Techsplainers explores the agent control plane, the centralized system that helps organizations deploy, operate, monitor and govern AI agents at scale. The episode explains the distinction between the control plane and the data plane, where individual agents perform tasks and interact with tools and data. It shows how a control plane provides a shared operating model for increasingly complex agent ecosystems by coordinating agents, applying policies, routing requests and maintaining visibility across enterprise environments.
The discussion examines key capabilities including access control, agent and tool registries, request routing, execution management, policy enforcement, state management, telemetry, lifecycle management and runtime orchestration. Listeners also learn how an agent control plane differs from Model Context Protocol, or MCP, and why the two address different layers of an agentic AI system. Finally, the episode explores the role of control planes in multi-agent workflows and how centralized governance, observability and orchestration can help organizations scale agent environments more reliably and securely.
Find more information at https://www.ibm.com/think/topics/agent-control-plane
Find all the Techsplainers episodes here: https://www.ibm.biz/techsplainers-podcast
Narrated by Derek Robertson

This episode of Techsplainers explores AI agent management, the practice of supervising, coordinating and governing AI agents as they become part of enterprise operations. The episode explains why managing agents becomes increasingly important as organizations move from isolated AI experiments to environments with dozens or even hundreds of specialized agents. It looks at the risk of AI agent sprawl, where disconnected agents operate across teams, tools and departments with inconsistent policies, limited visibility and fragmented oversight.

The discussion breaks down the core components of AI agent management, including governance and policy controls, human oversight, identity and access management, observability and lifecycle management. Listeners learn how management layers can coordinate agents, standardize access to enterprise systems, monitor behavior and support multi-agent workflows. The episode also examines how managing AI agents differs from traditional AI operations, where the focus was primarily on models and data pipelines, and why visibility into agent decisions, permissions and actions becomes critical as agentic AI scales.

Find more information at https://www.ibm.com/think/topics/ai-agent-management
Find all the Techsplainers episodes here: https://www.ibm.biz/techsplainers-podcast

Narrated by Derek Robertson

This episode of Techsplainers explores the EU AI Act, the European Union’s landmark law for regulating artificial intelligence through a risk-based framework. The episode explains who the act applies to across the AI value chain, including providers, deployers and importers, and why its reach can extend beyond Europe when AI outputs are used inside the EU. It breaks down the law’s core structure, covering prohibited AI practices, high-risk systems and the transparency obligations that apply to other types of AI use.
The discussion also examines what compliance looks like in practice. Listeners learn about requirements for high-risk AI systems such as lifecycle risk management, data governance, technical documentation, conformity assessments and post-market monitoring. The episode also covers the separate rules for general-purpose AI models, including added obligations for models considered to pose systemic risk. Finally, it connects the act to digital sovereignty, showing how AI governance, auditability and regulatory alignment are becoming part of broader business resilience strategies. Along the way, it also outlines the law’s penalties and phased implementation timeline.
Find more information at https://www.ibm.com/think/topics/eu-ai-act
Find all the Techsplainers episodes here: https://www.ibm.biz/techsplainers-podcast
Narrated by Amanda Downie

This episode of Techsplainers explores continuous compliance and how organizations Techsplainers explores continuous compliance and how organizations use automated, ongoing checks to stay aligned with regulations and frameworks instead of relying only on periodic audits. The episode explains how continuous compliance works in practice, from translating legal and policy frameworks instead of relying only on periodic audits. The episode explains how continuous compliance works in practice, from translating legal and policy requirements into measurable controls to collecting evidence across identity systems, cloud environments, configuration tools and vulnerability scanners. It also walks through how governance, risk and compliance platforms help validate that evidence, report on gaps and support remediation workflows.
Along the way, the discussion clarifies the difference between continuous compliance and continuous monitoring, showing how monitoring provides the raw signals while compliance maps those signals to regulatory obligations such as HIPAA, GDPR, SOC 2, ISO 27001 and PCI DSS. It also examines why continuous compliance matters for audit readiness, incident response, forensic analysis and cost control, especially as noncompliance can increase the financial impact of a breach. Finally, the episode connects continuous compliance to digital sovereignty, highlighting how real-time visibility and verifiable control can help organizations demonstrate trust, resilience and regulatory alignment.
Find more information at https://www.ibm.com/think/topics/continuous-compliance
Find all the Techsplainers episodes here: https://www.ibm.biz/techsplainers-podcast
Narrated by Amanda Downie