AI safety trust crisis: Individuals must lead public persuasion
Public trust in AI safety is collapsing—can lone voices rebuild it before AGI arrives?
The *LessWrong* article titled *Trust is Gone: AI Safety Needs Individuals* highlights a critical trust deficit that could undermine AI safety efforts. Authored by *warner*, the piece argues that humanity’s Preparedness Paradox—where preparations for hypothetical disasters are dismissed if no disaster occurs—has eroded public trust in institutions, leaving them ill-equipped to communicate AI risks effectively.
The essay identifies two key failures in public communication: Type 1, where the public misunderstands what was *known* before a disaster (e.g., COVID-19’s evolving guidelines), and Type 2, where efforts to *prevent* disasters (like Y2K) are dismissed as overhyped. The author notes that trust in U.S. institutions is at historic lows, complicating AI safety advocacy. With institutions struggling to rebuild credibility, the burden may shift to individuals—despite their instability—to persuade the public on AI risks before AGI arrives.
- Public trust in institutions is at historic lows (Pew Research), complicating AI safety advocacy.
- The Preparedness Paradox undermines AI safety efforts: preventive actions are dismissed if no disaster occurs, eroding credibility.
- Individuals—not institutions—may need to lead AI risk communication, as institutional trust is too fractured.
Why It Matters
Rebuilding public trust in AI safety is critical before AGI arrives—without it, even proven risks may be ignored.