# The Risk of Locked-In Values: Ideological Fanaticism and Technological Maturity

> Coverage of lessw-blog

**Published:** February 12, 2026
**Author:** PSEEDR Editorial
**Category:** risk

**Tags:** AI Safety, Existential Risk, Ideology, History, Future of Humanity

**Canonical URL:** https://pseedr.com/risk/the-risk-of-locked-in-values-ideological-fanaticism-and-technological-maturity

---

A recent analysis from lessw-blog examines the historical prevalence of ideological fanaticism in major atrocities and argues for its critical importance as a risk factor in the era of superintelligent AI.

In a recent post, **lessw-blog** discusses the existential risks posed by ideological fanaticism, particularly as humanity approaches the development of superintelligent AI and the state of "technological maturity." While much of the current discourse surrounding AI safety focuses on technical alignment—ensuring a system functions as intended without unintended consequences—this analysis pivots to the nature of the intent itself. It scrutinizes the "terminal preferences" that might be encoded into powerful systems, questioning what happens if those preferences are derived from fanatical worldviews.

The conversation regarding long-term future risks often centers on political structures, such as the dangers of totalitarianism or authoritarian surveillance. However, the author argues that focusing on **ideological fanaticism** provides a more precise and predictive framework. The post posits that fanaticism—defined by epistemic and moral certainty, extreme tribalism, and a willingness to use brutal violence—was the primary driver behind eight of the ten greatest atrocities since 1800. This distinction is vital because political systems are often merely the _means_ of execution, whereas ideologies represent the _ends_ or the terminal values that guide action.

This topic is critical for the PSEEDR audience because it bridges historical sociological data with futurist technical concerns. As we build systems capable of optimizing for specific outcomes with unprecedented power, the specific values guiding those optimizations become existential variables. The concept of "technological maturity" implies a point where technological capabilities stabilize, potentially allowing the dominant values of that era to be locked in indefinitely. If a superintelligent system is aligned with a fanatical ideology—one that claims absolute certainty and demonizes out-groups—the resulting "lock-in" could be catastrophic and permanent, effectively automating the destructive patterns of the past.

Furthermore, the post serves as a counter-narrative to the assumption that liberal democracy and rationalism are the inevitable end-states of history. The author notes that while fanaticism may seem less globally dominant today than during the peaks of the 20th century, it has seen a measurable resurgence in the last two decades. This trend suggests that the window for ensuring safe, pluralistic values in our technological infrastructure may be tightening, and that the "human element" of AI risk remains as volatile as the software itself.

We recommend reading the full analysis to understand the specific historical definitions used and the implications for future value alignment.

[Read the full post](https://www.lesswrong.com/posts/sFewetPnb8KJgcER7/long-term-risks-from-ideological-fanaticism-1)

### Key Takeaways

*   Ideological fanaticism is distinct from political totalitarianism and represents the "terminal preferences" of a group.
*   Historical analysis suggests ideological fanaticism drove 80% of the major atrocities since 1800.
*   The core characteristics of this risk are epistemic certainty, moral absolutism, and extreme tribalism.
*   As AI approaches superintelligence, there is a risk of permanently "locking in" fanatical values, making human ideological trends a technical safety concern.

[Read the original post at lessw-blog](https://www.lesswrong.com/posts/sFewetPnb8KJgcER7/long-term-risks-from-ideological-fanaticism-1)

---

## Sources

- https://www.lesswrong.com/posts/sFewetPnb8KJgcER7/long-term-risks-from-ideological-fanaticism-1
