Red vs blue: The parable of the feud within a feud
A fictional clan feud reveals why AI safety efforts may fail due to internal divisions...
Deep Dive
Joe Rogero's 'Red vs Blue' parable on LessWrong uses a clan feud to show how well-intentioned individuals may join a risky majority to protect loved ones, leading to catastrophic outcomes. The piece highlights the tension between individual loyalty and group wisdom.
Key Points
- The parable uses a clan feud to illustrate how individuals may join risky majority actions to protect loved ones, leading to disastrous outcomes
- It critiques the 'safety vs. progress' debate in AI alignment, highlighting how emotional bonds and social pressures can override rational risk assessment
- The piece references rationalist concepts like 'Moloch' and coordination problems, suggesting AI safety efforts may fail without better collective action mechanisms
Why It Matters
This parable reveals the psychological and social traps that could derail AI alignment efforts, urging better coordination.