Judges, Magistrate Judges, and Magistrates

23-1023.00

Median wage $153,990/yr24,030 employed (US)Rank #921 of 923 scored · top 100% by substitution

Arbitrate, advise, adjudicate, or administer justice in a court of law. May sentence defendant in criminal cases according to government statutes or sentencing guidelines. May determine liability of defendant in civil cases. May perform wedding ceremonies.

Sub-scores

0–100 · band = confidence interval from rater disagreement

Substitution4

Exposure5

Augmentation40

Substitution — the headline: capability discounted by cost, barriers and adoption.

Exposure — technical capability alone, regardless of whether anyone deploys it.

Augmentation — how much AI assists without replacing. High here + moderate substitution = a changing job, not a disappearing one.

Tasks on the substitution scale

21 rated tasks, binned by substitution score.

Position among all scored occupations

Distribution of 923 occupation scores; the marker is this occupation.

Tasks with substitution ≥ 70

Run 1.0.0-draft.1 · computed 2026-08-05 · rater panel: claude-sonnet-5, claude-haiku-4-5-20251001 · intervals span rater disagreement.

Why this score

The five weighted dimensions of the composite, averaged across this occupation's tasks (importance-weighted, panel mean). Exact weights and formulas: /api/v1/methodology.

Task automatabilityw 35%5

panel mean rating 1.2/5 → substitution pressure 5/100

Technical feasibility todayw 20%4

panel mean rating 1.2/5 → substitution pressure 4/100

Cost vs. human wagew 15%8

panel mean rating 1.3/5 → substitution pressure 8/100

Adoption barriersw 20%inverted — strong barriers lower the score1

panel mean rating 4.9/5 (barrier strength) → substitution pressure 1/100

Sector adoption velocityw 10%3

panel mean rating 1.1/5 → substitution pressure 3/100

Task breakdown (21 tasks)

Substitution pressure per task, weighted by O*NET importance in the composite. Expand a task for the full rater audit trail — every rating, every model, every rationale.

Read documents on pleadings and motions to ascertain facts and issues.

CI 24–37 · exposure 30 · augmentation 75 · importance 4.6/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Judicial systems are highly regulated and slow to adopt automation. While some courts and judges use e-filing and summary tools, they do not delegate the core reading and fact-finding to AI; adoption remains at the pilot and augmentation stage rather than replacement.
Sector adoption velocity	claude-sonnet-5	2/5	Courts are historically slow adopters of new technology due to procedural rules, ethics concerns, and caution around AI errors in legal contexts, though pilots for document review assistance are emerging.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI-powered document summarization, fact extraction, and issue highlighting can meaningfully assist judges in reviewing large volumes of filings, flagging relevant case law, and organizing key points—substantially raising their productivity while the judge retains final interpretive authority.
Augmentation potential	claude-sonnet-5	4/5	AI tools can efficiently highlight relevant facts, summarize lengthy pleadings, and flag key issues, meaningfully speeding up a judge's or clerk's review process while the judge retains final interpretive authority.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI can extract and summarize key facts and issues from legal documents, judicial reading requires nuanced interpretation of legal precedent, strategic positioning, and judgment about what is materially relevant—tasks that involve discretion and context beyond pattern recognition. Current systems cannot reliably perform this end-to-end with 50% time savings at equal quality.
Task automatability	claude-sonnet-5	2/5	AI can summarize and extract facts/issues from legal documents quickly, but reliable identification of legally salient issues at a level suitable for judicial decision-making requires nuanced legal judgment that current systems cannot fully replicate end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Constitutional and statutory law mandate that a judge (a licensed, appointed official) must personally read and rule on pleadings and motions; no AI system can legally supplant this duty. The liability and constitutional accountability remain with the human judge.
Adoption barriers	claude-sonnet-5	4/5	Judicial fact-finding and legal interpretation require an authorized judge exercising legal judgment; due process and legal authority requirements strongly limit full delegation to AI.
Cost vs. human wage	claude-haiku-4-5-20251001	4/5	Document processing and initial fact/issue extraction via AI is substantially cheaper than paying a judge or judicial officer to read and digest every filing in full, though oversight and verification remain necessary human work.
Cost vs. human wage	claude-sonnet-5	3/5	AI-assisted document review is cheaper than extensive human reading time, but the need for careful human verification of legal accuracy narrows the cost advantage.
Technical feasibility today	claude-haiku-4-5-20251001	3/5	AI document analysis and summarization tools exist in production (e.g., legal tech platforms, contract analysis engines), but they typically flag issues and extract facts rather than replacing the judge's interpretive role. Deployed systems have material error rates in understanding intent and legal sufficiency.
Technical feasibility today	claude-sonnet-5	2/5	Legal AI tools (e.g., document review, summarization software) exist and are used by clerks/attorneys, but no deployed product independently and reliably performs judicial-grade issue-spotting in production for judges.

Provide information regarding the judicial system or other legal issues through the media and public speeches.

CI 23–32 · exposure 25 · augmentation 63 · importance 2.9/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Judiciary systems are among the slowest sectors to adopt experimental AI for high-stakes communication. Precedent, institutional conservatism, and public trust constraints mean pilot use of AI for judicial messaging remains rare.
Sector adoption velocity	claude-sonnet-5	2/5	Courts and judiciary are generally slow to adopt AI tools, especially for public-facing communications where reputational risk is high.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI can significantly accelerate judicial communication by drafting clear explanations of legal principles, organizing talking points, and suggesting language while the judge retains full editorial control and final authority over the message.
Augmentation potential	claude-sonnet-5	3/5	AI can help judges research topics, draft talking points, and prepare speech outlines, meaningfully aiding preparation even though delivery remains human.
Task automatability	claude-haiku-4-5-20251001	2/5	AI can generate factual content about legal systems and produce speech drafts, but judges must exercise judgment about what to communicate, to whom, and with what institutional authority. The human role in selecting message, tone, and audience remains central.
Task automatability	claude-sonnet-5	2/5	AI can draft explanatory content about legal systems, but public speaking, media appearances, and answering unscripted questions require live human presence and judgment about what is appropriate to say as a sitting judge.ureau
Adoption barriers	claude-haiku-4-5-20251001	4/5	Judges hold institutional authority and public trust; judicial communications must bear the judge's signature and accountability. Ethical, reputational, and potential liability concerns create strong organizational and professional barriers to full automation.
Adoption barriers	claude-sonnet-5	4/5	Judicial ethics rules constrain what judges can say publicly about pending matters, and public communications from a judge carry authority and legitimacy that only the office-holder can provide.
Cost vs. human wage	claude-haiku-4-5-20251001	4/5	AI-assisted drafting of judicial communications costs a fraction of a judge's time to produce the same output; the inference and integration costs are negligible compared to the judge's hourly rate.
Cost vs. human wage	claude-sonnet-5	2/5	AI drafting assistance is cheap, but the actual public-facing delivery still requires the judge's time and cannot be substituted, limiting overall cost savings.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	While AI can draft explanatory legal content and speeches, no deployed system reliably performs the full task of a judge providing authoritative judicial communication. The reputational and legal stakes require human oversight that is not yet systematic in production.
Technical feasibility today	claude-sonnet-5	2/5	AI writing tools can help draft speeches or talking points, but no deployed product actually delivers judicial media appearances or public speeches for judges.

Research legal issues and write opinions on the issues.

CI 18–24 · exposure 25 · augmentation 63 · importance 4.5/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Judicial adoption of AI for opinion research is minimal; courts remain highly conservative institutions with deep institutional inertia, mandatory human judging authority, and zero tolerance for algorithmic displacement of core judicial functions. Adoption is confined to optional research tools, not opinion generation.
Sector adoption velocity	claude-sonnet-5	2/5	Courts are historically slow, heavily regulated, and cautious adopters of AI, with several jurisdictions issuing restrictive orders on AI use in legal drafting after citation-hallucination incidents, keeping deployment at the pilot/guidance stage.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI legal research and drafting assistants can help judges organize case law, identify relevant precedents, and outline arguments more efficiently, moderately enhancing research productivity. However, the augmentation is narrowly scoped to research and drafting support; the judge retains full analytical and decision-making responsibility.
Augmentation potential	claude-sonnet-5	4/5	AI tools meaningfully speed up legal research, case summarization, and initial drafting for judges and their clerks, substantially raising productivity even though the judge retains final authorship and judgment.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI can draft legal memoranda and outline arguments on straightforward issues, rendering binding judicial opinions requires fact-specific judgment, precedent synthesis, and authoritative legal reasoning that current systems cannot reliably produce end-to-end. The task demands error-free legal accuracy and constitutional legitimacy that AI cannot yet guarantee without substantial human oversight.
Task automatability	claude-sonnet-5	2/5	AI can draft summaries and surface case law quickly, but authoritative legal reasoning, weighing precedent, and final opinion-writing require judicial judgment that current systems cannot reliably replace at equal quality end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Judicial opinions are a core sovereign governmental function; judges are appointed or elected officials whose authority to render binding legal opinions is explicitly vested by constitutional and statutory law. No AI system can legally replace this decision-making role, and courts retain strict control over opinion authorship and publication.
Adoption barriers	claude-sonnet-5	5/5	Issuing binding legal opinions is a core judicial function requiring a duly appointed/elected judge; this is legally non-delegable and carries severe liability and due-process implications, representing a hard institutional barrier.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	AI legal research tools reduce research time but do not eliminate the need for judicial labor; integration costs, validation, and appellate risk management are high. The all-in cost of AI-assisted research remains comparable to or higher than the marginal cost of human judicial research, especially given error liability.
Cost vs. human wage	claude-sonnet-5	3/5	AI research assistance is cheap relative to clerk/associate time, but the judge's own analytical and drafting time, plus mandatory verification of AI-sourced citations, keeps overall cost roughly comparable rather than order-of-magnitude lower.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	AI legal research tools (LexisNexis, Westlaw AI) exist and assist with case law retrieval and summaries, but no deployed system independently writes binding judicial opinions at production scale. Experiments with AI-drafted opinions show material gaps in reasoning rigor and case law application, and no court system has authorized unsupervised AI opinion generation.
Technical feasibility today	claude-sonnet-5	2/5	Legal research tools (e.g., Westlaw AI, Lexis+ AI, CoCounsel) are deployed and used by attorneys, but no product independently produces judicial opinions in production; hallucination risk in citations remains a known problem even in current deployments.

Write decisions on cases.

CI 0–15 · exposure 8 · augmentation 50 · importance 4.6/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Judicial systems are highly regulated, risk-averse, and process-bound; adoption of AI for actual decision-writing is minimal and restricted to narrow administrative tasks, not case rulings.
Sector adoption velocity	claude-sonnet-5	2/5	Courts are historically slow to adopt new technology, constrained by procedural rules, ethics opinions, and institutional caution around AI in judicial reasoning.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI can assist with legal research, case summarization, and drafting templates, but augmentation is limited by judges' existing workflows and the narrow, high-stakes nature of decision-writing where human deliberation dominates.
Augmentation potential	claude-sonnet-5	4/5	AI can meaningfully assist judges by drafting summaries, researching precedent, and generating first-pass language, improving efficiency while the judge retains final authority and judgment.
Task automatability	claude-haiku-4-5-20251001	1/5	Writing legally binding case decisions requires nuanced interpretation of law, precedent, and facts—tasks demanding human judgment and accountability that current AI cannot perform end-to-end at the required quality and liability standard.
Task automatability	claude-sonnet-5	2/5	AI can draft portions of legal reasoning or summarize precedent, but crafting a legally binding judicial decision requires authoritative judgment, fact-finding, and accountability that current systems cannot fully replicate end-to-end at equal quality.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Judges are licensed officers of the court with constitutional and statutory authority to render decisions; only a human judge can legally sign and issue a binding ruling, creating an absolute legal barrier to automation.
Adoption barriers	claude-sonnet-5	5/5	Judicial decisions must be issued by a legally authorized judge; this is a core sovereign function with strict licensing, due process, and accountability requirements that cannot be delegated to AI.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	Even if AI drafting tools were deployed, judges' salaries and liability exposure are high; oversight and revision costs would likely exceed the human wage, and quality requirements preclude full automation.
Cost vs. human wage	claude-sonnet-5	2/5	While AI drafting could reduce time on research and boilerplate, the mandatory human review, verification, and legal liability keep the effective all-in cost close to or above the human cost for the actual decision task.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed product reliably writes complete judicial decisions in production; AI generates drafts with significant error rates, hallucinations on law, and lacks the legal authority and accountability required for binding rulings.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product autonomously issues judicial decisions in production; use is limited to research/drafting-assistance tools, not authoritative opinion writing.

Monitor proceedings to ensure that all applicable rules and procedures are followed.

CI 0–13 · exposure 8 · augmentation 38 · importance 4.7/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Judicial systems are conservative, heavily regulated, and slow to adopt automation. No meaningful production deployment of AI procedural monitoring exists across courts; adoption remains negligible.
Sector adoption velocity	claude-sonnet-5	1/5	Courts are notoriously slow to adopt AI for core adjudicative functions due to legal, ethical, and constitutional constraints, with adoption limited to administrative support tasks.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can assist judges by highlighting procedural rules, flagging deviations in filings, and surfacing prior rulings, meaningfully reducing research and documentation time while the judge retains decision authority.
Augmentation potential	claude-sonnet-5	2/5	AI can assist with transcript review, flagging procedural issues after the fact, or research support, but offers minimal real-time assistance during active proceedings monitoring.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI can parse legal documents and flag procedural discrepancies, the task requires real-time judgment about contextual compliance, discretionary exceptions, and dynamic courtroom conduct. Current systems lack the situated reasoning and authority to make binding procedural determinations.
Task automatability	claude-sonnet-5	1/5	Monitoring live courtroom proceedings for procedural compliance requires real-time judgment, authority, and legal interpretation that current AI cannot exercise as a decision-maker.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Judicial authority is statutorily and constitutionally vested in licensed judges; only a human judge can legally ensure procedural compliance and has duty of care. Substitution is legally prohibited regardless of AI capability.
Adoption barriers	claude-sonnet-5	5/5	This is a core judicial function requiring a legally authorized, licensed judge; constitutional and procedural law mandate human adjudication and oversight.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	AI systems for legal compliance flagging are moderately expensive to implement and maintain, while judicial salaries are high; cost advantage is marginal and does not offset the need for human oversight and correction.
Cost vs. human wage	claude-sonnet-5	1/5	There is no viable AI substitute performing this function, so no meaningful cost comparison exists; the human judge is irreplaceable in this role.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed products perform end-to-end procedural monitoring with legal authority. Research prototypes exist for document analysis, but no production system reliably substitutes for a judge's active oversight of live proceedings.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product performs this task; AI tools exist for legal research or transcription support but none monitor and enforce procedural compliance in live proceedings.

Perform wedding ceremonies.

CI 0–9 · exposure 8 · augmentation 13 · importance 2.4/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	This task has zero adoption of AI because legal and constitutional barriers make substitution impossible; no sector is attempting or considering AI-driven marriage ceremonies.
Sector adoption velocity	claude-sonnet-5	1/5	This niche ceremonial task within judicial duties shows no meaningful AI adoption trend; the judiciary broadly adopts AI slowly and this specific function is legally human-only.
Augmentation potential	claude-haiku-4-5-20251001	1/5	AI offers no meaningful productivity enhancement to the core act of performing a wedding ceremony, which is a formulaic legal pronouncement requiring human judgment only on whether the couple meets statutory requirements (which may be pre-screened outside this task).
Augmentation potential	claude-sonnet-5	2/5	AI could help draft or personalize ceremony scripts or vows, but offers minimal assistance to the actual act of presiding and legally solemnizing the marriage.
Task automatability	claude-haiku-4-5-20251001	1/5	Wedding ceremonies require legal authority, personal presence, and witnessed pronouncement of vows—elements that are fundamentally incompatible with AI automation regardless of efficiency gains. Current AI cannot legally solemnize marriages or fulfill the statutory role of the presiding official.
Task automatability	claude-sonnet-5	2/5	Officiating a wedding is a live, in-person ceremonial and legal act requiring physical presence, speech delivery, and signing of documents; AI cannot substitute for the human presiding officer today.atability is minimal beyond scripting.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Hard legal barriers prevent automation: statutes in every jurisdiction explicitly require a licensed judicial officer or notary to perform and sign the marriage certificate, and the official must be physically present and personally attest to the ceremony.
Adoption barriers	claude-sonnet-5	5/5	Marriage officiation is legally restricted to authorized individuals (judges, clergy, certain officials) whose signature makes the marriage valid, a hard licensing and legal barrier.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	AI cannot perform this task at all, making cost comparison meaningless. The alternative—hiring a licensed judge or magistrate—far exceeds any conceivable AI inference cost since the human is mandatory.
Cost vs. human wage	claude-sonnet-5	1/5	There is no viable AI substitute performing this task, so cost comparison favors the human by default; any AI cost would be additive to still needing a licensed officiant.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No AI system can legally perform or conduct a wedding ceremony in any jurisdiction; this task requires a human official with judicial or notarial authority present in person to sign documents and pronounce the couple married.
Technical feasibility today	claude-sonnet-5	1/5	No deployed AI product performs legal marriage officiation; this remains purely a human, legally-defined function with no automation in production.

Sentence defendants in criminal cases, on conviction by jury, according to applicable government statutes.

CI 0–0 · exposure 0 · augmentation 38 · importance 4.8/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	No sector is adopting AI to autonomously sentence defendants; judicial systems remain highly regulated and human-centered. This task is explicitly protected by legal requirement for human judicial authority.
Sector adoption velocity	claude-sonnet-5	1/5	The judiciary is a highly regulated, low-digitization-of-decision-making sector with strong resistance to automating actual adjudicative acts like sentencing.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI can assist judges by providing sentencing guideline calculations, prior case summaries, or risk assessments to inform discretionary decision-making, but such tools are limited in scope and the judge retains full decision authority.
Augmentation potential	claude-sonnet-5	3/5	AI can assist judges by summarizing case law, sentencing guidelines, and prior comparable sentences, improving consistency and research speed, but the final judgment remains human.
Task automatability	claude-haiku-4-5-20251001	1/5	Sentencing requires legal judgment, discretion within statutory bounds, and consideration of individual circumstances (prior history, remorse, victim impact). No AI system today can autonomously impose sentences that would be legally valid or defensible; this requires a licensed judge's authority.
Task automatability	claude-sonnet-5	1/5	Sentencing requires discretionary moral and legal judgment, weighing aggravating/mitigating factors, and personal accountability that current AI cannot legitimately perform end-to-end.'
Adoption barriers	claude-haiku-4-5-20251001	5/5	Sentencing is explicitly reserved for licensed judges by statute and constitutional law across all jurisdictions. Only a human judge with proper authority can lawfully impose a sentence; this is a hard legal barrier.
Adoption barriers	claude-sonnet-5	5/5	Sentencing is a core judicial function requiring a constitutionally and statutorily authorized judge; due process, appellate review, and separation-of-powers doctrines make this legally impossible to delegate to AI.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	The judicial authority to sentence cannot be cost-substituted; a judge must perform this task by law. Any AI assistance would only supplement, not replace, the judge's role, making cost comparison inapplicable.
Cost vs. human wage	claude-sonnet-5	1/5	There is no viable AI substitute performing this function, so cost comparison is moot; any AI assistance is a small fraction of a judge's total workload cost.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed product performs autonomous sentencing in any jurisdiction. Sentencing recommendations or analytical tools exist, but no system is legally authorized or actually performs the judicial act of imposing a sentence.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product actually issues criminal sentences; AI sentencing-support tools (e.g., risk assessment scores) exist only as advisory inputs, not as the decision-maker.

Rule on custody and access disputes, and enforce court orders regarding custody and support of children.

CI 0–0 · exposure 0 · augmentation 38 · importance 4.7/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Judiciary adoption of AI for core ruling is extremely limited; courts remain heavily process-bound with strong resistance to automation of judgment itself, reflecting both legal constraints and public expectation.
Sector adoption velocity	claude-sonnet-5	1/5	Court systems are notoriously slow to adopt automation for adjudicative functions, with strong institutional, ethical, and legal resistance to delegating rulings to AI.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI can assist with document review, legal research, and case organization, but the core judgment—weighing evidence, determining child welfare, ruling on disputes—must remain with the judge. Augmentation is limited to peripheral support.
Augmentation potential	claude-sonnet-5	3/5	AI can help judges by summarizing case files, researching precedent, and drafting portions of orders, but the core adjudicative task itself sees only moderate assistance.
Task automatability	claude-haiku-4-5-20251001	1/5	This task requires interpreting nuanced family circumstances, applying discretionary judicial judgment, and making child welfare determinations that fundamentally depend on human reasoning about complex social and emotional factors. Current AI cannot perform the judicial decision-making component end-to-end.
Task automatability	claude-sonnet-5	1/5	This requires exercising judicial discretion, weighing credibility, equity, and the best interests of children in fact-specific disputes, which is not something current AI can perform end-to-end at equal quality.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Hard legal barriers exist: only licensed judges with constitutional or statutory authority can rule on custody disputes and issue binding court orders. A human judge must personally perform this task by law.
Adoption barriers	claude-sonnet-5	5/5	Judicial rulings on custody require a duly appointed, licensed judge with legal authority and due process protections; this is a paradigmatic case of legally mandated human decision-making.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	The task involves irreplaceable human judicial authority with no feasible AI substitution path, making cost comparison moot; any AI support tools remain peripheral to the core judgment.
Cost vs. human wage	claude-sonnet-5	1/5	There is no viable AI substitute performing this judicial function, so no meaningful cost comparison exists; the human judge remains the only option.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed AI system reliably performs judicial rulings on custody disputes. Such decisions require legal authority vested only in human judges and involve liability and ethical responsibilities that AI cannot assume.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product issues binding custody rulings or enforces court orders; AI legal tools remain confined to research, drafting, or analytics support.

Instruct juries on applicable laws, direct juries to deduce the facts from the evidence presented, and hear their verdicts.

CI 0–0 · exposure 0 · augmentation 25 · importance 4.7/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Zero adoption of AI to replace judicial instruction and verdict hearing is occurring or likely, as the task sits at the core of constitutional due-process requirements and judicial authority that cannot be delegated to machines.
Sector adoption velocity	claude-sonnet-5	1/5	The judiciary is a highly regulated, low-digitization sector for core adjudicative functions, with essentially no movement toward AI performing this task.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI might assist judges by summarizing evidence or flagging case law, but the instruction itself, the exercise of judicial discretion, and verdict reception remain human-centric and non-augmentable in the strictest sense of the task statement.
Augmentation potential	claude-sonnet-5	2/5	AI may help draft jury instructions or summarize evidence beforehand, but it offers minimal real-time assistance during the actual instructing and verdict-hearing process.
Task automatability	claude-haiku-4-5-20251001	1/5	This task requires human legal authority, judgment about case-specific evidence interpretation, and interaction with jurors to ensure proper legal understanding. Current AI cannot legally instruct juries or make discretionary rulings on evidence admissibility and legal applicability.
Task automatability	claude-sonnet-5	1/5	Instructing juries and receiving verdicts is a core judicial function requiring live human authority, real-time adaptation, and legal legitimacy that current AI cannot replicate end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Hard statutory and constitutional barriers exist: judges must be licensed attorneys, appointed or elected under law, and the U.S. judicial system mandates human judges to hear cases, instruct juries, and render verdicts. Automation is legally prohibited.
Adoption barriers	claude-sonnet-5	5/5	This is a constitutionally and statutorily mandated judicial act requiring a licensed, appointed judge; no automation is legally permissible.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	The task cannot be automated at any cost because it requires human judicial authority; comparison of AI cost to human wage is moot when substitution is not legally or operationally possible.
Cost vs. human wage	claude-sonnet-5	1/5	There is no viable AI substitute performing this function, so cost comparison is moot; the human judge is the only legally functioning option.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed product can perform this task; it requires a licensed judge with statutory authority to conduct court proceedings, instruct juries, and render binding legal determinations. This is not a task any production AI system undertakes.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product performs jury instruction or verdict-taking in real courtrooms; this remains entirely outside current AI product deployment.

Preside over hearings and listen to allegations made by plaintiffs to determine whether the evidence supports the charges.

CI 0–0 · exposure 0 · augmentation 38 · importance 4.6/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Judicial systems are among the slowest to adopt automation due to legal mandates, public trust concerns, and the gravity of liberty/property consequences. No jurisdiction is replacing judges with AI decision-makers at scale.
Sector adoption velocity	claude-sonnet-5	1/5	Courts and judicial proceedings are a highly regulated, low-digitization-of-core-function sector with essentially no adoption of AI for presiding roles.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI can assist judges with research, case summarization, and document flagging, but these are peripheral to the core task of hearing testimony and rendering judgment. Augmentation potential is modest because the critical reasoning step remains exclusively human.
Augmentation potential	claude-sonnet-5	3/5	AI can assist with transcription, evidence summarization, legal research, and case preparation, aiding the judge's understanding of allegations even though it cannot preside itself.
Task automatability	claude-haiku-4-5-20251001	1/5	Judging requires legal interpretation, discretionary weighing of credibility and evidence, and application of contextual nuance that current AI cannot reliably replicate end-to-end. While AI can assist with document review or legal research, the core function of presiding over hearings and making determinative judgments on guilt or civil liability remains fundamentally human.
Task automatability	claude-sonnet-5	1/5	Presiding over live hearings requires real-time authority, discretion, and legal judgment exercised by an authorized officer of the court; no AI system can perform this end-to-end today.'
Adoption barriers	claude-haiku-4-5-20251001	5/5	Presiding over hearings and determining guilt/innocence is a legally protected function requiring a licensed judge or magistrate; statute and constitutional law mandate human adjudication with due-process protections that cannot be delegated to an automated system.
Adoption barriers	claude-sonnet-5	5/5	Judicial authority is a legally mandated, licensed role requiring constitutional and statutory authorization; only a sworn judge/magistrate can preside and rule, an absolute hard barrier.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	AI systems capable of legal reasoning and case evaluation remain expensive (high compute, specialized models), while judges are salaried across many cases, making the marginal cost per decision low. Full automation would require significant infrastructure investment without offsetting labor savings.
Cost vs. human wage	claude-sonnet-5	1/5	There is no viable AI substitute performing this function, so cost comparison is moot; any attempted automation would require extensive human oversight negating savings.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed judicial system relies on AI to independently preside over hearings or issue binding determinations of guilt/liability. Research prototypes exist for legal document analysis, but no production system performs the full task of hearing evidence and rendering judgment autonomously.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product presides over hearings or adjudicates evidentiary sufficiency; this remains outside current AI product scope entirely.

Rule on admissibility of evidence and methods of conducting testimony.

CI 0–0 · exposure 0 · augmentation 38 · importance 4.6/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Judicial systems are highly regulated and conservative; there is no meaningful real-world adoption of AI making admissibility rulings, and institutional and legal barriers prevent rapid change.
Sector adoption velocity	claude-sonnet-5	1/5	Judicial decision-making is highly resistant to automation due to legal, ethical, and due-process constraints, with essentially no production deployment of autonomous AI judges.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI might assist judges by summarizing case law or flagging precedents on evidence rules, but the core task—making the ruling itself—remains entirely human and cannot be materially augmented by current systems without reducing judicial authority.
Augmentation potential	claude-sonnet-5	3/5	AI can assist judges by summarizing precedent, flagging admissibility issues, or organizing evidentiary arguments, but the final ruling and reasoning remain human-driven.
Task automatability	claude-haiku-4-5-20251001	1/5	This task requires nuanced legal judgment, precedent interpretation, and real-time courtroom decision-making involving competing arguments about evidence and procedure. Current AI cannot reliably make binding evidentiary rulings that meet the threshold of equal quality and 50% time savings.
Task automatability	claude-sonnet-5	1/5	Ruling on evidence admissibility requires real-time legal judgment, weighing case-specific context, constitutional rights, and courtroom dynamics that current AI cannot reliably execute end-to-end.'
Adoption barriers	claude-haiku-4-5-20251001	5/5	Constitutional and statutory law explicitly require a licensed judge or magistrate to preside and rule on evidence. No automation can substitute for this legally mandated human role; liability and judicial authority are hard barriers.
Adoption barriers	claude-sonnet-5	5/5	This is a core judicial function requiring constitutional authority, licensing as a judge, and legal accountability; only an authorized human judge can make binding rulings.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	AI inference is cheap, but the human judge's role here is irreplaceable by cost alone—the judicial decision itself carries legal weight and liability that cannot be outsourced to an algorithm at any price point.
Cost vs. human wage	claude-sonnet-5	1/5	Since AI cannot perform this task at all, there is no viable cost comparison—human judges remain the only functional option.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed product performs this task reliably in production; ruling on evidence admissibility demands legal authority, courtroom presence, and accountability that AI systems do not possess. This remains research-stage.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product makes binding evidentiary rulings in courtrooms; AI legal tools remain research/assistive stage, not autonomous decision-makers.

Award compensation for damages to litigants in civil cases in relation to findings by juries or by the court.

CI 0–0 · exposure 0 · augmentation 50 · importance 4.6/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Judicial systems are among the slowest to adopt automation of core decision-making. Awarding damages is a central judicial function; bar associations, courts, and legislatures have shown minimal appetite for AI replacement of judicial judgment on liability and damages.
Sector adoption velocity	claude-sonnet-5	1/5	The judiciary is a highly regulated, slow-adopting sector for core decision-making functions, with AI use largely restricted to administrative or research support, not rulings.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can usefully assist judges by synthesizing precedent, calculating guideline ranges, summarizing case facts, and flagging comparable awards, raising efficiency in research and analysis. However, the final award remains the judge's independent decision.
Augmentation potential	claude-sonnet-5	3/5	AI can assist judges by summarizing case law, calculating damages ranges, or drafting portions of opinions, improving efficiency while the judge retains final decision authority.
Task automatability	claude-haiku-4-5-20251001	1/5	Awarding compensation requires legal judgment, interpretation of precedent, discretionary assessment of harm, and consideration of case-specific equities. Current AI cannot perform this end-to-end; it lacks the authority and judgment required to bind litigants to financial awards.
Task automatability	claude-sonnet-5	1/5	Awarding damages requires judicial discretion, legal authority, and case-specific judgment under law; no AI system today can perform this end-to-end task with equal quality at scale.
Adoption barriers	claude-haiku-4-5-20251001	5/5	This task is protected by hard barriers: only a licensed judge has the legal authority to award damages in civil cases. Constitutional and statutory law mandate that judicial decisions be made by human judges; liability for erroneous awards rests on the judge personally.
Adoption barriers	claude-sonnet-5	5/5	This is a core judicial function requiring a duly appointed judge or magistrate to exercise legal authority; statutory and constitutional requirements mandate human adjudication.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	The task requires a judge's full attention and responsibility. AI assistance tools cost far less than a judge's time, but replacing the judge entirely is not feasible; thus the relevant comparison is marginal cost of AI augmentation versus judicial labor, which remains expensive.
Cost vs. human wage	claude-sonnet-5	1/5	Since AI cannot legally perform this task, there is no valid cost comparison—human judicial officers remain the only permissible actor, making AI substitution cost irrelevant/infeasible.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed product performs judicial damage awards in practice. While AI can assist with prediction or analysis, no court or jurisdiction uses an AI system to independently determine and award compensation in binding civil cases.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product issues binding damage awards; AI tools at best assist with calculations or precedent research, not the authoritative act itself.

Conduct preliminary hearings to decide issues, such as whether there is reasonable and probable cause to hold defendants in felony cases.

CI 0–0 · exposure 0 · augmentation 38 · importance 4.6/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Judicial systems move slowly on procedural innovation and face structural constraints; preliminary hearings remain conducted by judges in all U.S. jurisdictions with minimal automation of the core decision-making function.
Sector adoption velocity	claude-sonnet-5	1/5	The judiciary is a highly regulated, low-digitization sector for core adjudicative functions, with essentially no adoption of AI to perform judicial rulings.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI could assist judges by summarizing evidence or flagging case law, but current systems offer limited augmentation on the core task of weighing probable cause testimony and making binding legal determinations.
Augmentation potential	claude-sonnet-5	3/5	AI can help judges by summarizing case files, legal research, and drafting preliminary notes, but the core adjudicative decision remains entirely human.
Task automatability	claude-haiku-4-5-20251001	1/5	This task requires legal judgment, interpretation of evidence under specific procedural rules, and constitutional authority that only a human judge can exercise. No AI system can legally make binding determinations of probable cause that affect a person's liberty.
Task automatability	claude-sonnet-5	1/5	Determining probable cause requires legal judgment, weighing witness credibility, and constitutional authority that current AI cannot exercise end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	5/5	This task is protected by hard legal barriers: constitutional law and court procedure require a licensed judge to conduct hearings and make probable cause determinations. Delegation to AI would violate due process and judicial authority requirements.
Adoption barriers	claude-sonnet-5	5/5	Only a licensed, sworn judicial officer can legally preside over hearings and make probable-cause determinations, a hard constitutional and statutory requirement.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	Judicial determination of probable cause requires a salaried judge; AI tools that might assist cost far less per task but cannot replace the judge, making direct cost comparison infeasible and replacement economically moot.
Cost vs. human wage	claude-sonnet-5	1/5	There is no viable AI substitute performing this function, so no meaningful cost comparison exists; human judges remain the only option.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed AI product can conduct preliminary hearings or make legally binding probable cause determinations. While AI can assist with document review, only licensed judges can perform this judicial function in practice.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product conducts preliminary hearings or issues judicial probable-cause determinations; this remains firmly outside any production AI system's scope.

Advise attorneys, juries, litigants, and court personnel regarding conduct, issues, and proceedings.

CI 0–0 · exposure 0 · augmentation 38 · importance 4.5/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	There is no meaningful adoption of AI to perform judicial advising, and institutional and legal barriers prevent any. Courts are not moving to displace judicial guidance with AI systems; any AI role is strictly supplementary to human judges.
Sector adoption velocity	claude-sonnet-5	1/5	Judicial functions are among the most conservative and heavily regulated, with essentially no production AI adoption for direct judicial advisement.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI can marginally assist judges by drafting research summaries or flagging procedural requirements, but the core task—authoritatively advising attorneys, juries, and litigants—remains the judge's sole responsibility and is not substantially enhanced by AI tools.
Augmentation potential	claude-sonnet-5	3/5	AI can help judges research precedent, draft procedural explanations, or prepare bench notes, offering moderate assistance while the judge retains full authority and delivery.
Task automatability	claude-haiku-4-5-20251001	1/5	This task requires nuanced legal judgment, contextual understanding of court procedure, and authoritative guidance that must account for individual circumstances and jurisdiction-specific rules. Current AI systems cannot reliably provide binding legal advice or procedural direction with the accountability required of judicial officers.
Task automatability	claude-sonnet-5	1/5	This requires real-time judicial authority, discretion, and legally binding guidance in live courtroom settings, which off-the-shelf AI cannot perform end-to-end today.
Adoption barriers	claude-haiku-4-5-20251001	5/5	This task has the strongest possible barriers: only a licensed, appointed judge or magistrate can legally advise on court proceedings and provide binding procedural guidance. Constitutional and statutory requirements mandate human judicial authority; no AI system can replace this function.
Adoption barriers	claude-sonnet-5	5/5	Only a licensed, constitutionally appointed judge or magistrate can legally advise on courtroom conduct and proceedings; this is a hard legal and licensing barrier.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	AI systems cannot legally substitute for judicial advice-giving, so cost comparison is moot. Where AI might assist (e.g., legal research), the overhead of human judicial review and sign-off means the total cost remains higher than simply having the judge advise directly.
Cost vs. human wage	claude-sonnet-5	1/5	There is no viable AI substitute performing this function, so no meaningful cost comparison exists; the human judge is irreplaceable for this role.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed product performs judicial advising at scale; this function is exclusively performed by licensed judges and magistrates who carry personal legal authority and liability. AI systems lack the legal standing and institutional authority to provide binding guidance in court proceedings.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product provides authoritative judicial advisement to attorneys or juries in actual proceedings; this remains outside current product scope entirely.

Grant divorces and divide assets between spouses.

CI 0–0 · exposure 0 · augmentation 38 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Judicial tasks are tightly regulated and bound to individual human judges; no sector-wide adoption of AI for issuing divorce decrees or asset orders is occurring or legally permissible today.
Sector adoption velocity	claude-sonnet-5	1/5	Court systems are notoriously slow to adopt automation for substantive rulings, especially those with binding legal and financial consequences like divorce decrees.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI can assist judges by summarizing case law, calculating asset valuations, or organizing discovery documents, modestly raising productivity in research and preparation. However, the core deliberation and judgment remain fundamentally human, limiting augmentation scope.
Augmentation potential	claude-sonnet-5	3/5	AI can assist judges by summarizing case facts, drafting asset division calculations, or flagging precedents, but the final judgment and decree remain human-driven.
Task automatability	claude-haiku-4-5-20251001	1/5	Granting divorces and dividing assets requires legal judgment, weighing competing interests, interpreting family law statutes, and making discretionary rulings that balance equity and fairness. Current AI cannot perform the core deliberative and authoritative function end-to-end.
Task automatability	claude-sonnet-5	1/5	Granting divorces involves exercising legal authority and judgment on equitable division of assets, which requires authoritative human adjudication and cannot be performed end-to-end by AI today.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Divorce and asset-division decrees must be issued by a legally authorized judge; statute and constitutional due-process requirements mandate human judicial authority and signature. Liability for erroneous division falls on the judge, creating legal barriers to substitution.
Adoption barriers	claude-sonnet-5	5/5	This is a core judicial act requiring a licensed, appointed judge with legal authority; statutes and constitutional due process mandate human adjudication and signature on the order.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	A judge's labor (including benefits and overhead) spans multiple tasks; the marginal cost of this task is embedded in judicial salary. AI tools that assist cost extra without displacing the judge, making the all-in cost higher than human performance alone.
Cost vs. human wage	claude-sonnet-5	1/5	There is no viable AI substitute product to compare costs against; the human judicial process is the only legally valid mechanism, making AI not a comparable alternative.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed AI system can legally grant a divorce or authorize asset division; these acts require a licensed judge's signature and authority. AI tools may assist with document review or asset calculation, but cannot substitute for the judicial act itself.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product issues legally binding divorce decrees or asset division rulings; this remains squarely a judicial function performed by courts.

Interpret and enforce rules of procedure or establish new rules in situations where there are no procedures already established by law.

CI 0–0 · exposure 0 · augmentation 38 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	This task cannot be adopted for automation because judicial functions are inherently reserved to human judges by law and constitutional design, regardless of sector digitization trends.
Sector adoption velocity	claude-sonnet-5	1/5	The judiciary is a highly regulated, tradition-bound sector with minimal AI adoption for substantive judicial decision-making, only slow pilots for administrative support.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI tools can assist judges by summarizing relevant precedent or analyzing procedural implications, but the core act of interpreting and establishing rules remains a human judicial function. Assistance is narrow and supportive rather than transformative.
Augmentation potential	claude-sonnet-5	3/5	AI can assist by researching precedents, summarizing analogous cases, or drafting procedural options, but the judge must independently interpret and decide, so assistance is meaningful but partial.
Task automatability	claude-haiku-4-5-20251001	1/5	This task requires judicial authority, legal discretion, and the ability to make binding decisions in novel procedural situations—functions that cannot be delegated to AI and are constitutionally vested in human judges. No current AI system can autonomously interpret law and establish precedent.
Task automatability	claude-sonnet-5	1/5	This requires exercising discretionary legal judgment in novel situations, weighing precedent, policy, and fairness—AI cannot reliably perform this end-to-end with equal quality today.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Judges are officers of the court operating under constitutional and statutory authority; only a human judge can legally interpret rules of procedure and establish new ones. Liability, authority, and legal requirement all prevent substitution.
Adoption barriers	claude-sonnet-5	5/5	Judicial authority to interpret and create procedural rules is a core sovereign function requiring a duly appointed, licensed judge with legal authority—automation is legally barred.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	The task requires a licensed judge with years of training; there is no cost-effective AI substitute for the human whose role is mandated by law and constitutional structure.
Cost vs. human wage	claude-sonnet-5	1/5	There is no viable AI substitute performing this function, so cost comparison favors the human judge by default; any AI assistance still requires full judicial oversight and salary.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed product performs independent judicial rule-making or procedure interpretation. While AI can assist with legal research, the core task of establishing binding procedural rules remains exclusively within judicial authority.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product autonomously establishes new procedural rules or renders binding judicial rulings; this remains firmly research/conceptual at best.

Participate in judicial tribunals to help resolve disputes.

CI 0–0 · exposure 0 · augmentation 38 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	No jurisdiction is replacing judges with AI systems. Courts remain among the most conservative and heavily regulated sectors, with strong institutional and legal resistance to automating the judicial function itself. Adoption of AI in courts is limited to research assistance, not decision-making.
Sector adoption velocity	claude-sonnet-5	1/5	The judiciary is a highly regulated, tradition-bound sector with minimal AI adoption for core adjudicative functions, despite some use in research/admin support.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI can assist judges with legal research, document summarization, and case law analysis, raising efficiency in preparation, but the core task of presiding over and resolving disputes remains human. The assistance is peripheral to the primary judicial function.
Augmentation potential	claude-sonnet-5	3/5	AI can assist judges with legal research, document review, and drafting support, improving efficiency, though the core deliberative and decision-making task remains human-led.
Task automatability	claude-haiku-4-5-20251001	1/5	Resolving disputes requires legal judgment, contextual interpretation of evidence, and discretionary sentencing decisions that depend on human authority and accountability. AI cannot perform the core judicial function of rendering binding verdicts, nor can it replace the human decision-maker whose legitimacy underpins the rule of law.
Task automatability	claude-sonnet-5	1/5	Judicial dispute resolution requires legal authority, discretion, and accountability that current AI cannot exercise; no off-the-shelf system can perform this end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	5/5	This task is protected by hard legal and constitutional barriers: only licensed judges and magistrates with statutory authority can preside over tribunals and render binding decisions. Judicial independence and human accountability in dispute resolution are legal requirements in virtually all jurisdictions.
Adoption barriers	claude-sonnet-5	5/5	Judicial authority is constitutionally and statutorily reserved to appointed/elected humans, with strict licensing, due process, and legal accountability requirements making substitution essentially prohibited.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	The task requires a licensed, salaried judge or magistrate; any AI system would require significant infrastructure, oversight, and maintenance while still necessitating human judicial oversight. AI could not reduce total cost because human authority remains mandatory.
Cost vs. human wage	claude-sonnet-5	1/5	Since AI cannot legally or practically substitute for the judicial role itself, there is no viable cost comparison—human judges remain the only option.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed AI system performs the role of a judge or magistrate in any jurisdiction today. AI tools exist for legal research and case analysis, but these are supplements to human judges, not replacements. The task requires statutory authority vested in specific individuals, which AI cannot hold.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product acts as a judge or tribunal member in production; AI legal tools remain research-stage or assistive only for this specific function.

Issue arrest warrants.

CI 0–0 · exposure 0 · augmentation 25 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	There is no adoption of AI warrant issuance because it is legally and constitutionally prohibited. Judicial sectors are not adopting automation for this core function, and no trend toward such adoption exists or could exist within current legal frameworks.
Sector adoption velocity	claude-sonnet-5	1/5	The judiciary is a slow-adopting, highly regulated sector, and warrant issuance specifically has seen essentially no AI deployment or displacement.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI can assist judges by summarizing facts, analyzing legal precedent, and organizing warrant application materials, but the core judicial decision and authority remain entirely human-driven. The assistance is preparatory rather than transformative of the judge's core role.
Augmentation potential	claude-sonnet-5	2/5	AI can help judges search precedent, summarize affidavits, or manage case files, but it provides minimal direct assistance to the actual probable-cause determination and signing act.
Task automatability	claude-haiku-4-5-20251001	1/5	Issuing arrest warrants requires legal authority, independent judgment of probable cause, and constitutional interpretation that only a qualified judicial officer can provide. No AI system can perform this task end-to-end today, as it involves authoritative legal decisions that must be made by a human with judicial standing.
Task automatability	claude-sonnet-5	1/5	Issuing an arrest warrant requires an exercise of judicial discretion and constitutional probable-cause determination that cannot be delegated to an AI system today, even though drafting or record-lookup portions could be assisted.
Adoption barriers	claude-haiku-4-5-20251001	5/5	This task has the strongest possible legal barrier: only a licensed judge or magistrate with proper judicial authority can lawfully issue arrest warrants under constitutional and statutory law. The human judicial officer is not merely preferred but legally mandatory.
Adoption barriers	claude-sonnet-5	5/5	Issuing warrants is a core judicial/constitutional function requiring a licensed judge or magistrate to make findings under oath; this is one of the strongest legal barriers to automation.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	AI cannot replace the cost of a judge's time because the task fundamentally requires a human judge's salary and legal authority. Any AI assistance would only reduce preparation time, not replace the judicial function itself, making the human cost unavoidable.
Cost vs. human wage	claude-sonnet-5	1/5	Because the task legally must be performed by an authorized judicial officer, there is no viable AI-only cost basis for comparison—human judicial cost is unavoidable.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed product performs judicial warrant issuance autonomously. While AI can assist in legal analysis, the actual act of issuing a warrant—a legally binding order—remains entirely within human judicial authority and is not delegated to AI systems in any jurisdiction.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product independently issues arrest warrants; this remains a judicial act requiring a magistrate's signature and legal authority, not a market for AI substitution.

Settle disputes between opposing attorneys.

CI 0–0 · exposure 0 · augmentation 38 · importance 4.2/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Judicial systems are among the most conservative and heavily regulated sectors; automation of core adjudication duties is explicitly prohibited by law and judicial ethics, resulting in virtually zero measured displacement.
Sector adoption velocity	claude-sonnet-5	1/5	The judiciary is a highly regulated, tradition-bound sector with minimal AI adoption for adjudicative functions, only slow uptake for administrative or research tasks.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI can assist with legal research, case summarization, and identifying precedent during case preparation, but it offers limited assistance during the core act of settling disputes—the judge's decision-making and ruling remain largely human.
Augmentation potential	claude-sonnet-5	3/5	AI can help judges by summarizing filings, researching precedent, and organizing arguments, but the actual dispute resolution and ruling remain human-led.
Task automatability	claude-haiku-4-5-20251001	1/5	Settling disputes between opposing attorneys requires nuanced legal judgment, interpretation of case law, evaluation of credibility and argument quality, and often discretionary rulings that depend on context, precedent, and equity—none of which current AI can perform reliably end-to-end as a replacement for a judge.
Task automatability	claude-sonnet-5	1/5	Settling disputes requires authoritative legal judgment, weighing credibility, procedural context, and binding decision-making that current AI cannot perform end-to-end with equal quality.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Settling disputes requires judicial authority vested in a licensed, appointed judge or magistrate by law; constitutional and statutory frameworks mandate a human judge sign off on the ruling, creating a hard legal barrier to automation.
Adoption barriers	claude-sonnet-5	5/5	This is a core judicial function requiring a licensed, sworn judge with legal authority; strict legal and constitutional requirements mandate human performance.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	A judge's or magistrate's salary (loaded cost often $100k–$300k+/year) is vastly higher than AI inference cost, but AI cannot perform the task at all, making the ratio comparison moot—AI is cheaper but not functional.
Cost vs. human wage	claude-sonnet-5	1/5	Since AI cannot legally or practically substitute for this task, there is no viable cost comparison—human judges remain the only real option.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed AI system can legally or reliably settle disputes between opposing attorneys. While AI can assist with legal research and document review, the final adjudication requires human judicial authority and accountability that no current product provides.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product independently mediates or rules on attorney disputes in courtrooms; AI is at most used for research or drafting support behind the scenes.

Impose restrictions upon parties in civil cases until trials can be held.

CI 0–0 · exposure 0 · augmentation 38 · importance 3.8/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Judicial tasks are among the slowest-adopting sectors for AI automation due to legal, constitutional, and accountability constraints. No jurisdiction has moved toward removing judges from restriction-imposition decisions.
Sector adoption velocity	claude-sonnet-5	1/5	Judicial decision-making authority sectors show extremely slow AI adoption for binding legal orders, with courts using AI only for research/drafting support, not decision authority.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI can assist by summarizing case facts, flagging precedents, or drafting restriction language templates, but the core act of imposing restrictions remains judicial discretion. Assistance is marginal relative to the judge's core responsibility.
Augmentation potential	claude-sonnet-5	3/5	AI can help judges research precedent, draft order language, and summarize case facts to inform restriction decisions, improving efficiency without replacing judicial judgment.
Task automatability	claude-haiku-4-5-20251001	1/5	Imposing restrictions on parties in civil cases requires nuanced legal judgment, understanding of case-specific circumstances, and discretionary authority that fundamentally depends on human judicial reasoning and accountability. Current AI cannot make binding legal orders or exercise the judicial discretion required.
Task automatability	claude-sonnet-5	1/5	Imposing legal restrictions (injunctions, restraining orders, bail-like conditions) requires authoritative legal judgment and formal court authority that current AI cannot exercise or execute end-to-end.'
Adoption barriers	claude-haiku-4-5-20251001	5/5	This task has absolute legal barriers: only a licensed judge or magistrate has statutory authority to impose court-ordered restrictions on parties. Judicial functions cannot be automated without fundamental legal change, and liability remains with the human judge.
Adoption barriers	claude-sonnet-5	5/5	Only a duly appointed judge or magistrate has legal authority to impose binding restrictions on parties; this is a core judicial power with strict licensing and constitutional constraints.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	A judge's compensation and liability are fixed by public budgets and law; AI infrastructure cost does not reduce the need for judicial oversight and signature. The task is not substitutable at lower cost because judicial authority cannot be delegated to AI.
Cost vs. human wage	claude-sonnet-5	1/5	There is no AI substitute performing this act at all, so cost comparison favors the human judge who is the only legally capable actor.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed product can independently impose legal restrictions in civil cases; this requires a licensed judge with statutory authority. AI can assist in drafting or analyzing restriction language, but cannot perform the authoritative act itself.
Technical feasibility today	claude-sonnet-5	1/5	No deployed AI product issues binding legal orders on parties in civil litigation; this remains purely a judicial function performed by humans.

Supervise other judges, court officers, and the court's administrative staff.

CI 0–0 · exposure 0 · augmentation 25 · importance 3.7/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Courts are among the most conservative institutions regarding automation, and judicial management functions are explicitly tied to human judicial office. There is no observable adoption of AI for supervising judges or court staff in production systems.
Sector adoption velocity	claude-sonnet-5	1/5	Judicial administration is a slow-moving, highly institutionalized sector with minimal AI adoption for supervisory or personnel management functions.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI could assist with data aggregation (caseload metrics, staff scheduling summaries) or draft administrative communications, but the core supervisory role—evaluation, mentoring, discipline, strategic planning—remains firmly human. Assistance is marginal and supplementary only.
Augmentation potential	claude-sonnet-5	2/5	AI can help with administrative tracking, scheduling, or performance data aggregation to support a supervising judge, but offers little assistance for the core interpersonal and authority-based supervisory judgment involved.
Task automatability	claude-haiku-4-5-20251001	1/5	Supervising human staff requires judgment, conflict resolution, performance evaluation, and organizational decision-making that depend on contextual understanding, emotional intelligence, and accountability—capabilities far beyond current AI systems. No AI today can autonomously manage personnel decisions, disciplinary actions, or strategic court administration.
Task automatability	claude-sonnet-5	1/5	Supervising other judges and staff requires personnel management, authority delegation, and interpersonal judgment that current AI cannot perform end-to-end; no off-the-shelf system can substitute for this managerial and authoritative role.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Judicial authority and supervision are legally vested in judges as officers of the court. Statutes, court rules, and constitutional principles require human judgment in personnel and administrative decisions; no AI can legally substitute for a judge's supervisory authority over court staff.
Adoption barriers	claude-sonnet-5	5/5	This is an inherently authoritative, legally-vested role requiring a commissioned judge; supervision of judicial officers is constitutionally/statutorily reserved to appointed or elected judges.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	The cost of AI infrastructure, oversight, and liability for managing judicial staff would far exceed the salary of a supervising judge. Judicial supervision carries legal and reputational stakes that make AI cost-prohibitive relative to human management.
Cost vs. human wage	claude-sonnet-5	1/5	There is no viable AI substitute performing this supervisory task, so cost comparison favors the human entirely; any AI cost would be additive, not substitutive.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed product performs judicial or court administrative supervision. This task requires human authority, legal accountability, and discretionary management decisions that have not been delegated to AI systems in any production court environment.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product performs judicial or court administrative supervision; this remains firmly a human institutional function with no AI product offering in production.

Related occupations — Legal

How to read this

A high substitution score does not mean this job disappears — it means a large share of its current tasks face replacement pressure, so the mix of tasks is likely to change. High augmentation alongside substitution typically means the occupation reorganizes around the protected tasks. Wide confidence intervals mean the rater panel disagreed: treat those scores as open questions, not verdicts.

What would change this score

New model capabilities (automatability, feasibility), falling inference costs (cost ratio), regulation and licensing shifts (barriers), and measured sector adoption (velocity) all re-enter at every index release. Each release is recomputed, versioned and kept queryable — scores are claims with a date on them, not permanent labels.