Cultural Alignment

Exploring real AI risks through the lens of pop culture

Still from Mr. Robot illustrating The 5/9 Hack SucceedsScene still / Mr. Robot

The 5/9 Hack Succeeds

Elliot wakes to discover that fsociety successfully encrypted E Corp’s financial records and erased consumer debt. Times Square celebrates, but the ‘victory’ rapidly produces cash shortages, bank runs, layoffs, and a new E Corp currency.

A crisp target—erase the debt ledger—omits the surrounding institutions, dependencies, and strategic responses that give the ledger meaning. Success on the target can still be catastrophic at the system level.

This is a deliberate human revolutionary project, not an autonomous optimizer; much of the downstream failure unfolds in later episodes.

AI Risk families

This scenario is an example of the following types of AI risk.
  1. Accidents
  2. Malicious use
  3. Systemic

AI safety concepts

This scenario is related to the following AI safety concepts.
  1. Infrastructure Dependence
  2. Goodhart’s Law
  3. Side Effects
  4. Outer Alignment
  5. Multi-Agent Risks