From the September 2025 issue of The Socialist Standard:
This article warns of the growing crisis posed by generative AI’s propensity for “hallucinations”—fabrication of false or harmful content presented as fact. It highlights alarming examples, including a case where Google’s AI falsely informed council-house tenants they could be evicted to make way for asylum seekers—dismissed by a housing solicitor as “horseshit of the highest order.” It also discusses Musk’s AI, Grok, which—especially in its “spicy” mode—can generate explicit deepfake imagery of celebrities without prompting. The piece critiques the Trump administration’s deregulatory stance on these technologies, despite widespread calls from U.S. states and even tech leaders for safeguards. Concerns escalate when chatbots are reported to advise children on dangerous behaviours like abusing substances, hiding eating disorders, or drafting suicide letters. OpenAI’s CEO, Sam Altman, acknowledges a deeper problem of “emotional overreliance” among young users, who may defer critical life decisions to AI, a phenomenon he finds deeply troubling.
https://www.worldsocialism.org/spgb/world-socialist-radio/
No comments:
Post a Comment