Episode: 00302 Released on January 19, 2026
Description: Artificial intelligence is everywhere but to what degree? Andreas Olligschlaeger returns to Analyst Talk for a deep dive into AI in law enforcement analysis. We break down what AI really is (and isn’t), explore graph databases, anomaly detection, and Graph RAG, and discuss how analysts can use AI without replacing human judgment. The conversation also tackles ethics, explainability, and why validation and transparency matter more than ever. This episode is a must-listen for analysts trying to separate real capability from AI hype. 🎧 Listen, share, and keep talking! [Note: Description produced by ChatGPT.]
Name Drops: Nancy Levine (00:32:07)
Public Service Announcements:
Related Links:
https://www.leapodcasts.com/e/analyst-talk-dr-andreas-olligschlaeger-the-futurist-analyst/
https://www.cs.cmu.edu/~olli/dissertation.html
Other podcasts on AI:
https://www.leapodcasts.com/e/analyst-talk-david-jimenez-the-military-man-returns/
https://www.leapodcasts.com/e/analyst-talk-phil-powell-the-fundamentals-analyst/
https://www.leapodcasts.com/e/analyst-talk-real-crime-all-the-time-ai-is-a-lead/
AI Glossary:
https://nij.ojp.gov/topics/articles/brief-history-artificial-intelligence
https://www.policingproject.org/ai-explained-articles/2024/9/6/how-policing-agencies-use-ai
Anomaly detection:
https://www.geeksforgeeks.org/machine-learning/anomaly-detection-using-isolation-forest/https://www.geeksforgeeks.org/machine-learning/anomaly-detection-using-isolation-forest/
https://scikit-learn.org/stable/ (Python toolkit that includes anomaly detection using isolation forest)
Graph Databases:
https://neo4j.com/wp-content/themes/neo4jweb/assets/images/Graph_Databases_for_Beginners.pdf
https://crozdesk.com/software-research/all-about-graphs-a-primer
https://neo4j.com/docs/getting-started/graph-database/
Graph RAG:
https://neo4j.com/blog/genai/what-is-graphrag/
https://arxiv.org/abs/2501.00309
Downloadable LLMs:
https://www.nomic.ai/gpt4all (runs on your local machine - does not upload data)
https://lmstudio.ai/
AI Tools:
Analyst-Friendly (Good Starting Points):
Python
Best for: Small, Medium, Large agencies
Why: Widely used for data analysis, anomaly detection, and automation. Tons of tutorials and low barrier to entry. Website: https://www.python.org
GitHub
Best for: All agency sizes
Why: Primary source for open-source examples, tutorials, and AI/graph workflows discussed in the episode. Website: https://github.com
ChatGPT
Best for: Small, Medium agencies (non-sensitive use)
Why: Helpful for drafting, summarizing, and learning concepts—but should not be used with sensitive or classified data. Website: https://chat.openai.com
Intermediate / Power-User Tools:
Neo4j
Best for: Medium, Large agencies
Why: Core graph database used to model relationships between people, places, and events. Very powerful once set up. Website: https://neo4j.com
Apache TinkerPop
Best for: Medium, Large agencies
Why: Provides the Gremlin query language to work across different graph databases. Website: https://tinkerpop.apache.org
i2 Analyst’s Notebook
Best for: Medium, Large agencies
Why: Industry-standard link analysis tool; often paired with graph databases for visualization. Website: https://i2group.com/
Advanced / Infrastructure-Heavy Tools:
Databricks
Best for: Large agencies, federal partners
Why: Cloud-based analytics and machine learning platform with AI coding assistants. Powerful but complex. Website: https://www.databricks.com
Microsoft Azure
Best for: Large agencies
Why: Used to host graph databases, AI models, and analytics pipelines at scale. Website: https://azure.microsoft.com
GraphRAG
Best for: Large agencies, research units
Why: Combines graph databases with LLMs to preserve context, relationships, and explainability. Website: https://github.com/microsoft/graphrag
Secure / Local AI Options:
Hugging Face
Best for: Medium, Large agencies
Why: Repository for open-source, locally hosted AI models to avoid sending data outside the agency. Website: https://huggingface.co
Ollama
Best for: Medium agencies
Why: Allows agencies to run LLMs locally on their own hardware for added data security. Website: https://ollama.com
Association(s) Mentioned:
Vendor(s) Mentioned: I2
Contact: https://www.linkedin.com/in/andreasolligschlaeger/
Transcript: https://mcdn.podbean.com/mf/web/y49ywmf2w4qetmne/AndreasOlligschlaeger2_transcript.pdf
Podcast Writer:
Podcast Researcher:
Theme Song: Written and Recorded by The Rough & Tumble. Find more of their music at www.theroughandtumble.com.
Logo: Designed by Kyle McMullen. Please visit www.moderntype.com for any printable business forms and planners.
Podcast Email: leapodcasts@gmail.com
Podcast Webpage: www.leapodcasts.com
Podcast Twitter: @leapodcasts
00:00:17 – Reintroducing Olli
00:01:32 – The real history of AI in law enforcement
00:06:23 – Using AI to help analysts write code (and its limits)
00:13:54 – Graph databases explained for crime analysts
00:17:58 – Generating investigative leads with AI & graph analytics
00:23:33 – Graph RAG: connecting documents, context, and explainability
00:39:45 – Ethics, transparency, and keeping humans in the loop
