Why AI Thinks Pikachu Is a Threat: Sound, Symbols & Survival Bias Podcast Por  arte de portada

Why AI Thinks Pikachu Is a Threat: Sound, Symbols & Survival Bias

Why AI Thinks Pikachu Is a Threat: Sound, Symbols & Survival Bias

Escúchala gratis

Ver detalles del espectáculo

Acerca de esta escucha

Is your favorite Pokémon secretly terrifying to AI? In this episode, we dive into a curious crossover of machine learning and sound symbolism, where algorithms trained on Pokémon names begin to overpredict threat. Researcher Alexander Kilpatrick uses XGBoost models across languages to test a wild theory: are these biases a form of digital cautiousness, just as error management theory predicts?

Tap play to unravel the unexpected ways AI perceives the world—and why it might think Pikachu packs a punch.

Love insights like these?
Download the ResearchBunny app or visit www.researchbunny.com to explore 10,000+ audio briefs just like this—across science, psychology, AI, and beyond.

adbl_web_global_use_to_activate_T1_webcro805_stickypopup
Todavía no hay opiniones