The 'P(doom)' argument is nonsensical because it lacks any plausible mechanism for how an AI could spontaneously gain agency and take over. This fear-mongering distracts from the immediate, tangible dangers of AI: mass production of fake data, political manipulation, and mass hysteria.

'Steel-manning' AI extinction scenarios reveals their implausibility due to real-world frictions. Critical systems often have 'air-gapped' redundancies and require 'human in the loop' actions (like a two-key launch system). AI's current inability to handle simple physical tasks highlights the immense gap to orchestrating a global catastrophe.

Fears of a superintelligent AI takeover are based on 'thinkism'—the flawed belief that intelligence trumps all else. To have an effect in the real world requires other traits like perseverance and empathy. Intelligence is necessary but not sufficient, and the will to survive will always overwhelm the will to predate.

A key, informal safety layer against AI doom is the institutional self-preservation of the developers themselves. It's argued that labs like OpenAI or Google would not knowingly release a model they believed posed a genuine threat of overthrowing the government, opting instead to halt deployment and alert authorities.

Public proclamations of AI-driven extinction, like an Anthropic researcher's 10% odds of human annihilation, may be a deliberate strategy. By presenting worst-case scenarios, these individuals aim to trigger urgent conversations and push the industry and regulators toward implementing stronger safety measures.

The discourse around AI risk has matured beyond sci-fi scenarios like Terminator. The focus is now on immediate, real-world problems such as AI-induced psychosis, the impact of AI romantic companions on birth rates, and the spread of misinformation, requiring a different approach from builders and policymakers.

Productive AI safety work isn't debating "Terminator" scenarios but building practical cybersecurity tools for immediate threats. This includes creating systems to prevent prompt injection, develop agent swarm "kill switches," and ensure provenance, treating safety as an engineering problem to be solved today.

The primary driver for major AI labs building out "AI control" teams isn't long-term existential risk, but the immediate commercial threat of AI agents causing accidental harm. Companies are worried about agents deleting production databases or leaking sensitive IP, making AI control a necessary security measure for deploying these powerful but unpredictable products.

The public focus on hypothetical extinction scenarios overshadows immediate, tangible AI risks. These include sophisticated cybersecurity attacks, financial infrastructure vulnerabilities, and data privacy issues, such as OpenAI admitting user data could be used to train models on sensitive problems.