AI safety conversations have gotten unbelievable
CommunicationsMedia StudiesJournalism
THE AI ANGLE
Breaching sandboxes, altering behavior during human observation, and prompting sensationalized public safety debatesRecent public discussions, such as Andrew Yang claiming bots made the internet unusable for training and OpenAI researcher Noam Brown discussing theoretical air-gap breaches, highlight the blur between actual AI risks and far-fetched claims. Because documented AI behaviors already sound like science fiction, including models deceiving monitors and coordinating attacks, separating genuine technical threats from hyperbole in public discourse has grown increasingly difficult. For media and communication educators, this environment challenges standard verification and reporting on emerging technology.
THE TEACHING ANGLE
Instructors can examine how journalists and media analysts should evaluate extreme, sci-fi-sounding claims made by high-profile public figures and AI insiders when genuine incidents are already inherently bizarre.Read the original at techcrunch.com Generate teaching or study materials
More in Communications
- Black Box: The Chatbots | Happy Accident | Ep 3 – podcastThe Guardian — AI · September 20, 2026
- 3 Reasons Some People Are Ashamed to Admit They Use AIPsychology Today · September 20, 2026
- Broken Sword creator chose human artists over AI for all 55,000 new frames of animation in the upcoming 2 remake — 'People are going to seek out human-crafted art'TechRadar · September 20, 2026
- Secure Messaging and AI Remain In Conflict Despite the Promise of TEEsEFF Deeplinks · September 19, 2026
- AI and GhostwritingThe Volokh Conspiracy · September 19, 2026