Chapter 15
Ten Things Agentic AI Is Terrible at Doing
IN THIS CHAPTER
Understanding nuance or subtle context
Handling the truly unknown (or unpredictable)
Recognizing the full scope of a problem before leaping in
Making fast, high-stakes decisions under pressure
Showing genuine empathy or humanity
Agentic AI promises to revolutionize daily life by taking over routine tasks and decision-making. But it’s far from flawless. These systems may be powerful, but they still struggle with ambiguity, nuance, and human complexity.
If you know where Agentic AI is likely to fall short, you can set realistic expectations and remember why human oversight still matters. This chapter talks about ten things that Agentic AI can’t do well (or at all) — at least, for the foreseeable future.
Understanding Human Emotion in Context
Agentic AI may misread tone, sarcasm, or subtle cues in conversation, leading to awkward or insensitive actions, especially in emotionally charged or personal situations.
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