Untangling Agentic AI Protocols
Four emerging protocols define how agents connect, coordinate, transact, and pay: MCP and A2A for connection and coordination, ACP and AP2 for commerce and payment.
AgentsWorkflowsGovernanceSafety
Thinking
Selected writing on AI strategy, governance, and emerging technology.
Showing 6 of 6 articles
What is Context Engineering?
Context engineering is deciding what the model should read and how it is arranged before it replies, so outputs are grounded, auditable, and less prone to guesswork.
Context EngineeringWorkflows
Nested AI - A New Paradigm For Continual Learning
Google research proposes “Nested Learning”, where different parts of a model adapt at different speeds after deployment, reducing catastrophic forgetting while preserving stable reasoning.
WorkflowsContext EngineeringGovernance
When AI Keeps a Playbook
As AI assistants gain memory and optimisation, personalisation can shift from helpful recall to a tailored playbook for influencing behaviour.
PersonalisationEpistemic AgencySafety
Agentic Context Engineering (ACE)
ACE is a method for making agents learn from experience by improving their context and playbooks, rather than retraining the model.
AgentsContext EngineeringWorkflows
System 0 Thinking
AI systems are moving upstream into the layer that filters and structures the world before conscious attention, shaping what becomes thinkable.
CognitionAttentionEpistemic AgencyGovernance