Chapter 7
Considering Risks, Ethics, and Hard Questions
IN THIS CHAPTER
Retaining basic human competency
Deciding who’s in control
Making Agentic AI transparent and explainable
Instilling AI systems with integrity
Avoiding rogue AI at the wheel
Chapter 10 covers all the bases in designing Agentic AI with safety in mind, but design is only half the challenge. After an autonomous system begins operating in the real world, it inevitably collides with messy human contexts and ethical gray zones. Inevitably, questions arise that aren’t purely technical: Who’s accountable when an agent makes a harmful decision? How do you balance efficiency against fairness, or innovation against long-term risk? And how do you prepare for unintended consequences that may emerge only after widespread adoption?
Building safe autonomy means asking uncomfortable questions early and being willing to design and redesign systems in light of them. This chapter focuses on those deeper considerations. It examines ...
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