{
  "id": 2940791,
  "title": "How, when and why to use agentic AI in our neuroscience labs",
  "url": "https://urgent.news/2026/08/24/how-when-and-why-to-use-agentic-ai-in-our-neuroscience-labs",
  "topic": "ai",
  "section": "AI",
  "published": "2026-08-24T04:00:42.000Z",
  "source": {
    "name": "The Transmitter",
    "slug": "the-transmitter",
    "url": "https://www.thetransmitter.org/artificial-intelligence/how-when-and-why-to-use-agentic-ai-in-our-neuroscience-labs/"
  },
  "original_language": "en",
  "account": "In early March, while on the Caribbean island of Barbados, Konrad Kording demonstrated how agentic AI could independently build web applications. This display of speed and capability highlighted the need for neuroscience labs to establish policies and practices for using AI. Kording emphasized that agentic AI affects three core aspects of neuroscience labs: research outcomes, skill development among lab members, and established cultural and methodological norms.\n\nDuring their retreat three weeks later, lab members quickly discovered how agentic AI could accelerate the development of new analytical pipelines. This led to a formal lab AI policy, as concerns emerged about AI potentially diminishing opportunities for deep, time-consuming skill-building among students and the uncertainty surrounding AI output verification.\n\nKey principles of the lab AI policy include prioritizing human skill development, verifying AI-generated results, avoiding risks by limiting AI access to necessary data folders, investing time in learning AI tools, and maintaining clear authorship and responsibility guidelines. By implementing these principles, the lab aims to navigate the complexities of integrating agentic AI into their work while safeguarding the integrity of their research and the development of their scientific community.",
  "summary": "With agentic AI’s rapid infiltration into science, researchers need to develop AI use policies and practices.",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}