The 'AI Safety' Movement Is Making AI Less Safe (reason.com)

Recent revelations from within major AI companies like Anthropic and OpenAI have sparked significant concern about the safety of artificial intelligence systems. Employees from Anthropic, including high-ranking members, have voiced alarming predictions, with some estimating a 10% chance that AI could lead to human extinction within the next decade. This has been linked to the "AI safety" movement, which seeks to regulate AI development under a narrative shaped by a philosophical framework known as TESCREAL (Technology, Effective Altruism, Social Responsibility, Existential Risk, and Alignment). Critics argue that this approach may inadvertently create a monopoly on AI knowledge, hindering open research and discussion necessary for safe AI usage. The implications of this movement are profound. By concentrating AI development within a few corporate giants and government entities, the potential for oversight and understanding decreases, leaving society vulnerable to the very risks these safety measures aim to mitigate. Advocates for democratizing AI technology argue that broader access could foster resilience against malicious uses of AI, using real-world examples where open-source solutions have proven effective in cybersecurity. As the debate unfolds, it raises critical questions about who controls AI technologies and how their governance could shape or diminish our future safety and autonomy.