Chapter 10
Building Agentic Systems Responsibly
IN THIS CHAPTER
Designing safe AI autonomy
Understanding what design principles to put in place
Training AI agents effectively
Using world models to train AI
Agentic systems are the next frontier of artificial intelligence (AI). These systems have immense potential, from accelerating scientific discovery to creating limitless new ways to shape the human experience. But with this power comes a heightened responsibility for developers, designers, and users.
Building agentic systems responsibly means more than just complying with regulations or minimizing risk. It requires embedding ethical foresight into design, establishing mechanisms for transparency and oversight, and creating pathways for human accountability at every stage of deployment.
Building responsibly also means recognizing that these systems need to interact with one another, with people, with human institutions, and with the broader digital ecosystem. Because of these interactions, you must make interoperability, safety, and trust central to the systems’ designs, rather ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access