Reflecting on the 2025 I3 (Intelligence – Innovation – Imagination) Conference

FBI AIA
5 days ago
3 min read

Written by FBI AIA Board Member, Nikki Rutman

I3’s second annual conference was held last month in the iconic International Spy Museum in Washington, DC. The event space, located at the top level, had exceptional views. The space was warm and inviting with intimate tables designed to foster conversation.

With a packed house of outstanding speakers, there was little time to chat between sessions. The rapid-fire agenda gave the audience 15-to-20-minute bites of information, enough to tickle the brain and tee up topics for further discussion. By midday, there were equal number of participants in the coffee space as were in the event space – symbolic of the crowd’s desire to engage in meaningful conversations driven by the conference’s content.

There were many key takeaways, and I would encourage you to attend next year’s event and submit a proposal. The networking alone is critical. Our own FBI AIA member, Dr. Charles Russo, did an exceptional job of discussing ethical frameworks for AI implementation. It would be great to continue to see the AIA represented at future iterations!

Key Takeaways:

Intelligence

We need to continue to frame the problem for decision makers, especially at a time when the world order is changing. We can no longer rely on the way the world used to be or how we want it to be. Rather, we must look at the world we are in and provide context for decision makers.
Think of the world as a system and remember to look up from the microscopic view to see how/if your findings are still relevant from a macro perspective.
Intelligence professionals must prove our relevance and develop trust among key stakeholders. The best way to do that is to be in the room with decision makers when the decisions are made.

AI

If you are building an in-house generative AI capability, you must leverage APIs and internal data into your model; it will make the outcomes more relevant to you and your stakeholders.
Create a framework to measure “success” from the outset. Benchmarking will be critical. You will inevitably be asked how you are using AI and what efficiencies it has created, so be ready.
AI is only as good as the question you ask of it.
Be the philosopher: ask where does a simple task end and the human-in-the-loop begin? Question whether AI is replacing thinking and the impact of cognitive offloading on you, your analysts, your stakeholders, etc.

Great Debate: AI - Friend or Foe?

AI might make you quicker, but is it effective? Could the result just be mediocre? There could be a happy medium: reduction of alert fatigue, potentially making issue identification faster to the customer. Humans would need to drive the process and validate.
Do we truly understand the motivation and intent of AI companies? Have we done threat assessments and due diligence checks? There are challenges between AI companies depending on whether they are publicly traded or not.
Will AI really replace intelligence analysts? It may replace analysts that only regurgitate the news. Analysts who embrace AI may replace those who don’t.

Humanity

The digital era puts human connection in jeopardy. Data doesn’t drive insight; it adds to it. Human fluency, understanding people, is very important and a dying skill. Be really informed but also be really connected.

Professional Development

A recent survey indicated that networking is very important. Of the people who took the survey, only 35% applied for a posting where the remainder had success with networking.
The perception of government experience has changed. There is an increasing demand for government experience.
There is also a cautionary tail of groupthink among intelligence managers. There is a need to leave room for real entry level analysts.