Skip to content

GCRI · Intelligenza artificiale e sistemi autonomi

IA e servizi pubblici

Governments use algorithms to target benefits, detect fraud, triage health care, allocate inspections, translate and answer citizens’ questions. When these sistemi are wrong, opaque or biased, people lose services and rights they depend on, often without knowing why. This page sets out how public sector AI succeeds and fails, how it is assessed and made accountable, and what evidence serves the agencies, auditors and citizens involved.

01Canvas

Come si svolge il lavoro.

sistemi · intorno a questo temaIllustrativo

Registro dei modelli

AI and public services · Model registry · Release gate

Un modello bloccato al gateGCRI · Intelligenza artificiale e sistemi autonomi

Modelli

M-843 triage assistantServizio pubblicoAlto22 of 24 suitesBloccato
M-266 flood nowcastScienzaMedio11 di 12Rilasciato
M-756 document searchInternoBasso8 of 8Rilasciato
M-769 agent plannerAgenticoAlto12 of 16In valutazione

Scheda del modello

Uso previsto
supporto al triage, decide una persona
Fuori ambito
decisioni di ammissibilità
Limiti noti
minore accuratezza sui casi rari
Titolare
Responsabile del modello

Registro degli incidenti

  1. D-13INCCitazione allucinata nel riepilogo
  2. D-6MITRetrieval grounding aggiunto
  3. D-2INCPrompt injection tramite file caricato
  4. D-1MITAutorizzazione dello strumento ristretta

Evaluation scores

vs release threshold

Task accuracy87 of 85 requiredRobustezza77 of 75 requiredFairness gap (lower is better)5 of 5 requiredRefusal on misuse94 of 95 required
Model registry: the screen’s structure, terms and states for AI and public services.Nomi di esempio e valori illustrativi; un progetto, non un registro né un sistema attivo.

Fig.Model registry: example names and illustrative values, not a record or a live system.

02La domanda

AI and public services: how the risk develops.

Public accountability, access and rights. The dossier below sets out how the risk develops, what is measured, where the data come from, how it is modelled and who decides.

Meccanismo

Errors fall on the vulnerable

People who depend most on public services are often those most affected by automated decisions, and least able to challenge them.

Meccanismo

Historical data carry bias

Models trained on past decisions can reproduce discrimination present in those decisions, even when protected characteristics are excluded.

Meccanismo

Opacity blocks challenge

When people cannot see that an algorithm was used or why it produced a result, they cannot contest it, and errors persist.

Meccanismo

Accountability stays with the agency

Procuring a system from a vendor does not transfer responsibility. Agencies must be able to explain, audit and override algorithmic decisions.

03Metodi ed evidenze

How AI and public services is modelled, measured and tested.

Il modello, le sue ipotesi, i suoi dati e la sua incertezza, nella forma in cui uno scienziato o un ingegnere li verificherebbe. Le strutture seguono metodi pubblicati. I valori rappresentati sono illustrativi, non risultati.

Modello di sistema

The couplings between the systems in AI and public services: signs of influence, a reinforcing and a balancing loop, and cross-links where stress jumps.

++−+Intelligenza artificialeDatiComunitàEconomiaR di rinforzoB bilanciamento
Causal loop diagram from the page’s systems; signs show the direction of influence. Illustrative structure for a system dynamics or network model, not estimated couplings.
Principi guida FAIR · GO FAIR e FORCE11W3C PROV · World Wide Web ConsortiumGuidance on consistent treatment of uncertainties · IPCCCEOS Analysis Ready Data · Comitato sui satelliti per l'osservazione della Terra
04Tra i domini

AI in public services touches the domains people depend on.

Public service AI studies are designed for the serie Evidence Brief. The EU AI Act classifies many public-service uses as high-risk; UNESCO’s Recommendation (2021) and WHO guidance (2021) set ethical expectations.

  • Algorithmic impact assessments a body publishes.
  • Accounts from people affected by automated decisions.
05Indicatori

Cosa viene misurato e con quale accuratezza.

Ogni indicatore riporta la sua unità, la sua cadenza, il tipo di fonte su cui si basa e l'etichetta di evidenza che un valore riporterebbe. Non sono mostrati valori: un programma li fornisce sotto la propria autorità.

IndicatoreUnitàCadenzaTipo di fonteEtichetta
Public algorithm registerscount of sistemi registeredper updateGovernment algorithm registersStatistiche pubblicate
Algorithmic impact assessments completedpercent of deployed systemsannualeRegistri delle agenzieDICHIARATO
Error rate by population groupdifference in error ratesper auditAlgorithm auditsOSSERVATO
Decisions appealed and overturnedcount and percentannualeAppeals recordsStatistiche pubblicate
Systems with human review of adverse decisionspercent of systemsper revisioneAgency documentationDICHIARATO
Audit findings on public sector algorithmscount by severityper audit cycleAudit office reportsRIPORTATO
06Fonti

Da dove provengono le evidenze.

Set di dati pubblici e registri istituzionali specifici, ciascuno con la sua classe di licenza. Le fonti soggette ad accordo sono utilizzate solo alle relative condizioni e mai ripubblicate.

FonteCosa fornisceClasse di licenza
Government algorithm and AI registers, such as the UK Algorithmic Transparency Recording Standard recordsPublic records of algorithmic tools in useOpen, Open Government Licence where applicable
National audit office reports on algorithmsIndependent audits of public sector AIPublic records, licence per office
Osservatorio OECD.AI sulle politicheGovernment AI policies and the AI Incidents MonitorPublic, OECD terms
Court and ombudsman decisions on automated decision-makingLegal findings on public sector algorithmsRegistri pubblici
Canada Directive on Automated Decision-Making and its impact assessment toolA model for impact assessment requirementsPublic, Government of Canada terms
07Decisioni

Chi decide, e a cosa servono le evidenze.

Studio Intelligenza artificiale e sistemi prepares evidence. The decisions below are taken by the authorities named, under their own rules, and never by the platform.

Chi decideLa domandaEvidenze fornite da questa pagina
Government agenciesWhether and how to deploy an algorithm in a public serviceImpact assessments; audit methods
Auditors and ombudspersonsWhere to investigate algorithmic harmRegisters; appeals data; audit evidence
ParlamentiWhat transparency and accountability to requireComparative policy and case evidence

Output prodotti da questo argomento

  • Impact assessment templates for public sector AI
  • Audit methods for disaggregated error and fairness
  • Transparency record templates
  • Case studies of public sector algorithm failures and fixes
08Questioni centrali

What public-sector technologists ask.

  • Is the system accurate for the whole population it serves, including minority groups?
  • Can affected people understand a decision, challenge it and get it corrected?
  • Who is accountable when an automated decision is wrong?
  • Does automation improve access, or does it create new barriers for people without digital skills or connections?
09Meccanismi

How automated decisions go wrong.

sistemi trained on historical administrative data inherit its biases and gaps. Risk-scoring tools can concentrate scrutiny on groups already over-represented in past enforcement. Small error rates become large numbers of wrongly affected people at national scale. When staff defer to the system, known as automation bias, errors are not caught.

Opacity compounds harm: if neither public servants nor citizens can see why a decision was made, it is hard to contest and slow to correct. Access barriers arise when services move online without alternatives.

10Metodi e dati

Evidence for accountable systems.

  1. 01

    Valutazione d'impatto

    Structured assessment of a system’s purpose, data, affected groups and safeguards before deployment, using scienze sociali e comportamentali. Quality depends on honest completion.

  2. 02

    Fairness and error analysis

    Disaggregated error rates by group, with probabilità e statistica. Different fairness definitions can conflict, so trade-offs must be made explicit.

  3. 03

    Explanation and contestability

    Designing decision notices and appeal routes people can use. Technical explanations do not by themselves make a decision contestable.

  4. 04

    Audit and monitoring

    Ongoing checks of outcomes and complaints after deployment.

12Where this page’s references

Where this page’s references lead.

Gli organismi e i documenti alla luce dei quali si legge questa pagina. Stato ed edizioni cambiano; ogni link porta all'organismo emittente.

13Domande

Domande e risposte.

What is automated decision-making?
Decisions about people made wholly or partly by algorithms, such as eligibility for benefits or prioritisation for inspection, which some laws regulate specifically.
What is an algorithmic impact assessment?
A structured review before and during deployment of an algorithm’s purpose, data, risks to rights and safeguards, often required for public sector systems.
Can people challenge an algorithmic decision?
Many legal systems give rights to an explanation, human review or appeal, but these depend on knowing that an algorithm was used and on how the law applies.
Should public services use AI at all?
That is a decision for public authorities and the public. The platform’s role is to show what systems can do reliably, where they fail and what safeguards evidence supports.
What is automation bias?
The tendency of people to over-rely on automated recommendations, accepting them without adequate scrutiny even when they are wrong.
Can affected people use the platform’s work?
Yes. Outputs are written to be usable by advocates, ombudspersons and the public as well as by public servants.
Does the platform evaluate specific government systems?
Only through scoped projects with the competent authority under written terms. It does not audit systems on its own initiative or publish assessments without agreement.
RapportiContribuisci

Contribuisci con evidenze da questa pagina.

Rapporti is the Consortium’s quarterly publication. What you know about this subject can become evidence, a case or a correction in the next edition, credited to you and published under CC BY 4.0 where it is your own work.

GCRIIntelligenza artificiale e sistemi autonomiTecnologia esponenzialeEvidenze di Labs
Intelligenza artificiale e sistemi autonomi

Porta un metodo, un set di dati o una domanda.

Informazioni e richieste sono aperte oggi. Una richiesta riceve un riferimento e indica il passo successivo; candidature, pagamenti, assegnazioni di calcolo e registrazioni si aprono solo quando esistono il team, le condizioni e la capacità di revisione.