Who is accountable when decisions are jointly produced by a user, an organization, a model provider, and an autonomous agent?
Responsibility gaps and 'moral crumple zones' — where blame lands on the nearest human operator rather than the system's designers — predate AI agents and get sharply worse with them. The question here is the human and organizational one; legal liability and remedies are treated under Power 4.1.
View on the map → · Open in Browse →
The responsibility debate sharpened around a practical worry: that nominal human review 'launders' liability, redirecting accountability from a system's designers to whoever clicked approve. One paper reframes the gap as 'consequence reception' — harm occurs, the system is identified, but no persistent agent absorbs consequences in a behavior-changing way — and concludes high-stakes agents must stay tethered to human principals until that architecture exists.
harm occurs, the producing system is identified, yet no continuing agent receives consequences in a way that changes future behavior.
Reframes the responsibility gap as 'consequence reception' and concludes high-stakes agents must stay tethered to human principals with meaningful control and proportional liability until an entity that can absorb consequences exists.
AI cannot behave responsibly. Only natural persons can.
The software-testing community's forceful statement of the human-accountability position: the operator of an agent always bears responsibility, because machines cannot participate in society's responsibility mechanisms.
The accountability that should flow toward the design and deployment of the system is quietly redirected to the person who clicked 'approve.'
Names the organizational mechanism — 'liability laundering' — by which nominal human review converts designer accountability into blame for the nearest reviewer; a corporate-practice update of the moral crumple zone.
Authors writing on the topic have employed a range of different, incompatible definitions, which are often imprecise, misleading, or otherwise inadequate.
From the literature's leading responsibility-gap skeptic, a conceptual cleanup arguing the debate rests on incompatible definitions — the current state of the philosophical argument, now under pressure from actual agent deployments.