\u201cEach and every day, we sit down as a group, we figure out the guardrails we\u2019re going to apply to these agents, [and] we ask ourselves: Is it risk that a human should be answering, or is it risk that an agent can answer?\u201d Eubank said. \u201cAnd right now, today, humans are all responsible for risk. We have not turned any agents loose to assume risk on their own behalf.\u201d<\/p>\n
The balance, he said, constituted a \u201cdelicate dance\u201d because the agents move so much faster than humans.<\/p>\n
\u201cBut,\u201d Eubank added, \u201cthis is a monumental point as we operate, secure, defend, and attack the network.\u201d<\/p>\n
Eubank told DefenseScoop on the sidelines of the annual TechNet Augusta conference that the task force is composed of about 8-10 civilians and service members, and is led by a lieutenant colonel that is soon set to promote.<\/p>\n
He said members of his command traveled to the West Coast and \u201cvisited with a bunch of really smart folks,\u201d though didn\u2019t detail who those people were during the interview, but referenced a trip to the Pentagon\u2019s innovation hub in California in an earlier speech. They came away from the trip having learned that the best way to integrate AI into the command was to form a small group and make the effort its sole job.\u00a0<\/p>\n
Since then, the task force is \u201cway ahead\u201d of where Eubank thought it was going to be, having created and trained agents that mimicked human work flows in about 45 days.<\/p>\n
\u201cWhether it\u2019s simple things like helping write sit reps, or it\u2019s getting after agentic red teaming or agentic mission elements, if somebody wants to do something, they go to them, and they\u2019re the ones that work it,\u201d he said. Members of his command can request help from the task force for AI-related projects, and the new formation will also keep tabs on its development.<\/p>\n
To the audience at the conference, Eubank said the task force trained agents \u201cto the same standard that humans get trained,\u201d adding that they get tasked with a mission that is overseen by humans.<\/p>\n
\u201cWhen they make a mistake,\u201d he said, \u201cthey get taught the way to do it and it goes back in and re-trains itself.\u201d\u00a0<\/p>\n
Eubank said the command has \u201cpurposely\u201d avoided using commercial frontier models, to show the work they\u2019ve been doing \u2014 which is still supported industry \u2014 can be done without having to go to larger companies.\u00a0<\/p>\n
He said the move away from those models was informed by the cost of tokens and a broader conversation about the lack of AI governance, adding that without it, \u201cwe\u2019re going to price ourselves out of business.\u201d<\/p>\n
![\"Drew](\"https:\/\/www.europesays.com\/ai\/wp-content\/uploads\/2026\/08\/1787187732_569_Drew-Lawrence-headshot.png\")
<\/p>\n