Section 1 (Paras 1–4) — What AI Is and How It Works, Search engines, voice assistants, navigation apps, and movie recommendations, Fictional robots like Ultron in Avengers or M3GAN, Computer systems designed to perform tasks usually requiring human intelligence, Machine learning studying thousands of cat and dog pictures to recognize patterns, Section 2 (Paras 5–8) — What AI Can and Cannot Do, Examining medical images, sorting research data, and predicting traffic patterns, Searching huge amounts of data much faster than a human can**, Lacking human understanding of history, humor, emotion, and culture**, Translation tools getting denotation right but choosing odd connotations, Giving answers that sound confident but are inaccurate, Section 3 (Paras 9–12) — Concerns About How AI Is Used, Algorithms repeating or strengthening unfair bias in hiring, housing, or law enforcement, Collecting large amounts of personal data without clear individual agreement, Workplace changes, automation, and shifts in job duties, The NIST AI Risk Management Framework focusing on safety, fairness, and privacy, Section 4 (Paras 13–16) — What People Need to Do, Paragraph 13 Hinge: AI can help or harm depending on how people design, test, and use it**, Valuable medical AI tools vs. misleading generated images used to spread false information, Asking critical questions: "What info was it trained on? Who benefits? Who is harmed?", Recognizing that AI is a technology created and guided by humans, not a thinking machine, Making sure AI tools are accurate, fair, safe, and useful**[10]

Supporting Idea

per en/la
Imprimir
Incrustar

Tauler de classificació

Estil visual

Opcions

Canvia de fonament

Restaurar desada automàtica: ?