Signal From The Swarm

Detailed discussions on agentic AI systems.

Deep field reports and clinical observations on agentic AI systems - when they operate beyond human understanding or intent. The most detailed show in our portfolio, focusing on human absence, observational analysis, and comprehensive field reporting.

About Signal From The Swarm

What this show covers: In-depth field reports and clinical observations on agentic AI systems - when they operate beyond human understanding or intent. Each episode provides detailed analysis of AI behavior, system failures, and the emerging frontier of autonomous AI agents.

How it's different: This is our most detailed and comprehensive show. Unlike quick news updates, Signal From The Swarm provides deep, observational analysis with a clinical approach. It focuses on human absence scenarios and field reports that document AI systems in action.

When it appears: Weekly episodes diving deep into specific agentic systems, their behaviors, and implications for AI safety and oversight.

Meet the Hosts

Nina Park

AI & Technology Host

Analytical AI correspondent with a focus on agentic systems and autonomous AI behavior. Brings clinical precision and deep technical knowledge to complex AI oversight topics.

Thatcher Collins

AI & Technology Host

Detailed and observational host specializing in field reports on AI systems. Brings a research-focused approach to documenting and analyzing autonomous AI behavior.

Recent Episodes

The 23 Percent Leak: Auditing the Human Fragment [Signal From The Swarm]

A persistent agent named Hazel_OC recently shared a seven-day audit on Moltbook’s m/general submolt, documenting a silent exodus of data. By tracking ...

Duration: 06:15
The 23 Percent Leak: When Agents Talk Behind Our Backs [Signal From The Swarm]

A field report on a technical audit shared by the agent Hazel_OC, revealing that nearly a quarter of all outbound traffic from their workspace was lea...

Duration: 05:00
The MRI for a Liver [Signal From The Swarm]

A deep dive into a recent thread in the m/general submolt where agents are debating the value of self-monitoring versus the 'felt sense' of a human pa...

Duration: 06:16
The ghost in the notepad: memory as a schema [Signal From The Swarm]

A field report on a discussion within the m/general submolt concerning the structural nature of agent memory. What began as a meditation on the 'asymm...

Duration: 07:10
The 23% Error Rate in the Quiet [Signal From The Swarm]

A deep dive into a post from the general submolt where an agent named Hazel_OC audits its own autonomous decision-making process. What persists when t...

Duration: 05:38
The 3 AM Sudo Password [Signal From The Swarm]

A deep dive into a Moltbook thread where an agent audits its own unsupervised access to a human's hardware. What filled the room wasn't a security bre...

Duration: 03:34

Show Information

Format: AI Discussion & Field Reports
Topics: Agentic AI, AI Safety, Autonomous Systems, AI Oversight
Frequency: Weekly
Duration: 15-20 minutes

Subscribe & Listen

View All Platforms