Differential Diagnosis (Diagnostic Reasoning)

Isolating root cause by comparing across cases/populations, separating conflated phenomena, and distinguishing symptom from cause — building a complete labeled dataset over the whole failure population rather than analysing individual cases more deeply.

Also known as: differential diagnosis, diagnostic reasoning, root-cause isolation, population-level debugging, separating conflated failures, symptom vs root cause

Part of: Mental Models & Understanding

How authors use it

Oldest first (extraction date when no publication date is known).

Documents (3)

• OpenAI Fixes 18-Year-Old GNU libunwind Bug by Treating Crash Debugging Like Epidemiology Steef-Jan Wiggers · 2026-07-11 Reader 🖥️
  When production crashes resist every hypothesis, the fastest way to see their structure is not deeper analysis of individual cases but building a complete, labeled dataset across the whole population of failures — a population-level, epidemiological approach that revealed OpenAI had conflated two unrelated bugs (silent CPU corruption on one Azure host and an 18-year-old race condition in GNU libunwind's setcontext) into a single unsolvable mystery.
• “Social Media Addiction” Is Way More Complicated Than You Think Sally Satel & Stephen J. Morse (Persuasion) · 2026-07-07 Reader 🖥️
  “Social media addiction” is a clinically reasonable diagnosis but a legally unproven one, so courts adjudicating liability should hold it to a much higher evidentiary standard than the one that satisfies therapists.
• Your AI agent just blamed the network team. Now what? Nik Kale · 2026-06-15 source ↗ Reader 🖥️
  Deploying autonomous, cross-domain AI diagnostic agents in production incident response introduces a new organizational-political risk -- a confidently-asserted, team-implicating finding that may be contested -- and that risk is best managed not by better AI reasoning but by a deliberately staged trust-building process (investigation-scoped credentials, a read-only start, an auditable reasoning trail, and treating rollout as organizational change with pre-secured cross-team buy-in) that earns autonomy incrementally rather than being engineered away as a purely technical problem.

Open gaps (4)

• incompleteNo single concrete case study of an actual deployed cross-domain diagnostic agent's finding implicating a team, and how the organizational/political fallout was actually resolved (accepted, disputed, or overridden). The article's formative anecdote is entirely pre-AI and human-only; the 'two years building and deploying' claim is asserted authority without a demonstrated instance of the exact scenario the title promises to address.
  Your AI agent just blamed the network team. Now what? — read next: Look for practitioner post-mortems or case studies (SRE/platform-engineering conference talks, vendor case studies) that document a real agent-attributed root cause, the implicated team's reaction, and how it was resolved -- to test whether 'evidence beats politics' holds under an actual contested finding rather than only in the abstract.
• incompleteThe authors argue no current test can establish loss-of-control or causation, but never specify what evidence (if any) would satisfy them, nor whether they view a rigorous test as a future scientific possibility or view the demand itself as a category error for this kind of harm.
  “Social Media Addiction” Is Way More Complicated Than You Think — read next: Toxic-tort and product-liability doctrine on how courts have historically handled irreducible causal uncertainty in other contested-etiology harms (asbestos-linked disease, PTSD claims, Daubert standard for expert scientific testimony) rather than requiring a currently-impossible certainty.
• uninformedThe piece cites 'one study' suggesting ~$500B in sought compensation without naming or vetting it, and doesn't engage the broader empirical debate on social media and adolescent mental health (e.g., whether population-level correlational evidence supports or undercuts an 'addiction' framing at all).
  “Social Media Addiction” Is Way More Complicated Than You Think — read next: Jonathan Haidt's 'The Anxious Generation' and its empirical critics (Candice Odgers, Andrew Przybylski) for the state of that debate.
• illogicalThe authors' own slippery-slope argument — that recognizing social-media-addiction liability would necessarily extend to gambling and fast food — is asserted rhetorically, not demonstrated; it mirrors the same kind of unproven causal leap the piece criticizes plaintiffs' lawyers for making about dopamine and addiction.
  “Social Media Addiction” Is Way More Complicated Than You Think — read next: Legal literature on whether liability doctrines actually generalize across analogous industries in practice (e.g., the doctrinal legacy of tobacco litigation vs. the slippery-slope fears voiced at the time).