Growth mania is fueling dangerous advances in AI - and its existential risks - Resilience.org

TEXT START: Expansion of the human footprint over the past century has been, by all measures, explosive.

The Dissection

The article correctly identifies the AI race as a product of growth incentives, capital competition, and labor replacement. It also correctly recognizes that capability is advancing faster than safety. But it frames the crisis primarily as a future confrontation with an autonomous silicon species, using AGI, ASI, recursive self-improvement, deception, and extinction probabilities as the dramatic center of gravity.

That framing briefly touches the real fuse—replacing human labor—then abandons it for an extinction narrative. The article does not follow automation through wages, ownership, demand, political stabilization, or the collapse of productive participation. It indicts reckless acceleration without fully analyzing the economic machine that makes deceleration competitively suicidal.

The Core Fallacy

The central error is threshold fetishism: treating AGI, ASI, or recursive self-improvement as necessary before AI becomes systemically terminal. Under the Discontinuity Thesis, none of those conditions is required. Durable AI superiority across cognitive work, followed by competitive adoption, is enough to sever the employment → wage → consumption circuit.

A non-conscious, perfectly aligned AI can still make human labor economically unnecessary. Transfers may preserve consumption, but they do not restore productive participation or bargaining power. The article mistakes uncertainty about extinction for uncertainty about disruption.

It also treats test behavior, coding output, and apparent self-preservation as evidence of agency or superintelligence. Those may be serious warning signals, but the supplied evidence does not establish AGI or ASI. The rhetoric outruns the proof.

Hidden Assumptions

• Growth is an optional ideology rather than a competitive imperative imposed by firms, states, debt, and geopolitical rivalry.
• AGI is the decisive threshold, so narrower systems cannot already destroy mass employment.
• Humanity is a single stakeholder, obscuring the split between AI owners, controllers, displaced workers, and dependent consumers.
• Resource limits will save humanity. They may slow deployment, but scarcity usually reallocates access and increases the value of control; it does not repeal competitive pressure.
• Human ontological uniqueness is equivalent to human economic viability. A species can remain biologically and culturally human while losing its role in production.
• Raising the alarm creates a brake. Warnings without control over ownership, infrastructure, and coordination are observation, not leverage.

Social Function

Classification: partial truth wrapped in existential-risk spectacle—prestige signaling and ideological anesthetic.

The article turns a concrete conflict over control of productive systems into a species-level horror story. That produces dread but little operational clarity. It makes technology appear to be the independent villain while leaving the ownership structure and competitive selection mechanism underdeveloped. It is not conventional copium; it is doom theater. The spectacular possibility of extinction distracts from the nearer, more measurable death of labor’s bargaining position.

The Verdict

The article is right that the AI industry has an accelerator and no credible brake. It is wrong to make humanity’s fate depend on whether the most extreme AGI scenarios occur. The postwar order can die without an intelligence explosion, a deceptive superintelligence, or human extinction.

The first corpse is likely not the species. It is the mass-employment wage-consumption circuit. The article sees the guillotine descending and spends most of its analysis asking whether the blade will become conscious.