Skip to content
NetworkingOnlinePast event

From Identity-First to Authority Continuity

Event artwork for From Identity-First to Authority Continuity

This event has ended

This listing is kept as an archive. Explore upcoming events to find your next one.

See upcoming events
About this event

What to expect

Traditional security architecture is fundamentally broken. We've built entire systems around identity, tokens, and role-based access control — and capability theorists have known for decades that this approach was flawed. The rise of AI agents has now made those cracks impossible to ignore, creating an urgent need to rethink how we model authority and access.

This online meetup, hosted by LFDT Kyiv, tackles one of the most important architectural shifts in modern security: moving from identity-centric models to authority continuity. Rather than treating security as something you possess (a token, a credential), this session explores authority as a continuous property of execution that can only shrink, never expand.

What You'll Explore

The session begins with first principles reasoning about what identity, identifiers, and authority actually are. You'll examine why common vulnerabilities like confused deputy attacks and privilege escalation aren't edge cases — they're the structural outcome of building systems on possession-based models.

Drawing on decades of research by Alan Karp, the meetup works through real-world delegation cases that break every IAM system built today:

  • Delegation and attenuation across organizational boundaries
  • Chained delegation involving multiple parties
  • Revocation scenarios, including the complex case of revoking access from someone you never directly delegated to
  • The confused deputy problem (known since 1988, still appearing in every generation of systems)
  • Seven aspects of sharing that people rely on in practice but that no identity-based, role-based, or attribute-based system handles correctly

Detailed use cases are available in Alan Karp's research, providing concrete examples of where current approaches fail.

Authority Continuity: A New Model

The second half explores how authority continuity fundamentally changes security architecture. Instead of starting with "who are you," this approach anchors authority at the origin of an intent, flows it through a verifiable chain, and ensures it cannot be expanded, stolen, or reinterpreted at any point in the execution path.

This model isn't theoretical — AI agents are accelerating the need for it. When autonomous agents act on behalf of users, traditional identity-first models create dangerous ambiguities about who authorized what action. Authority continuity provides

Running an AI event?

List it here, free.

Meetups, workshops and conferences go in front of thousands of builders after a quick review.

List your event