AI safety case
A structured, auditable argument that a system's residual risk is acceptable.
What it commits you to
A safety case is the structured argument that the claimed safety is supported by evidence, with assumptions stated so that a regulator can find the weak joint. It borrows from nuclear and aviation licensing, where it works because there is an independent regulator. Without that independence it degrades into an internal document, which is what most published cases currently are.
Which realms lean on this
Procedure, not conviction. This is the vocabulary of constitutions, audits and the people who have to sign.
governance · 2 realms — The Contractual Court, The Ubuntu Weave.
Canon that discusses this
declaration · 1986
The Asilomar AI Principles
Drafted over a few hours on the final day of a four-day 1986 conference on future technologies, and the first written set of research principles for advanced AI agreed by the researchers themselves. Its structure — general principles plus separately stated capability thresholds — is the structure of every frontier safety framework since. The excerpt below paraphrases the principles rather than reproducing the agreed wording.
primary · 300 BCE
Mozi
The one ancient Chinese text that argues for impartial care as a principle rather than as a sentiment: regard others' parents as your own, reject aggressive war and defensive hierarchy, and justify both by an explicit standard. The excerpt below paraphrases the jian xiang ai chapters rather than quoting a translation.
modern · 2019
Human Compatible: Artificial Intelligence and the Problem of Control
Russell argues that perfect value alignment is the wrong goal: systems are built by imperfect people, for imperfect objectives, and certainty about human values would be dangerous if it were achieved. The alternative is enough uncertainty about human preferences to keep a human in the loop as a genuine check. The excerpt below paraphrases the argument rather than quoting the 2019 book.
See also
External auditing
governanceIndependent verification of a developer's claims by an accredited third party.
Frontier safety framework
governanceA developer's published promise of what it will do at each capability threshold.
Institutionalisation
governanceA principle hardened into a rule enforced by an organisation that outlives its authors.
AI incident reporting
governanceMandatory disclosure of system failures so that others can learn from them.