Skip to content

GCRI · Inteligência artificial e sistemas autônomos

Cibersegurança da IA

AI and cybersecurity meet in three ways: AI sistemas can be attacked, AI can help attackers, and AI can help defenders. Models can be poisoned, stolen, manipulated or made to leak data; their supply chains of data, weights and libraries carry new risks. This page sets out how these threats work, how AI systems are secured, and what evidence serves the security teams and regulators responsible.

01Registrar

AI cyber security on the record.

02A pergunta

AI cyber security: how the risk develops.

Adversarial methods and model exposure. The dossier below sets out how the risk develops, what is measured, where the data come from, how it is modelled and who decides.

Mecanismo

Models are attack surfaces

Adversarial inputs, prompt injection, data extraction and model theft target the model itself. Classic application vulnerabilities still apply around it.

Mecanismo

The AI supply chain is long

Training data, pretrained weights, libraries and plugins come from many sources. Poisoned data or tampered models can introduce hidden behaviour.

Mecanismo

AI lowers the cost of some attacks

Generative AI can draft phishing, translate lures and help write code, lowering the skill barrier for some attacks. The size of this effect is studied and debated.

Mecanismo

AI assists defence

Machine learning detects anomalies, triages alerts and finds vulnerabilities. Defenders must also secure the AI tools they adopt.

03Simulador

Uma cascata através de redes interdependentes.

A pegada de perigo elimina os ativos colocalizados em cada camada; a fragmentação, a dependência entre camadas e a sobrecarga dos sobreviventes propagam depois a falha. A vista 3D separa as camadas por profundidade. Modelo didático com valores ilustrativos declarados, não uma previsão.

Controles

Ativos por camada
18
Área de impacto da ameaça
8% da área
Ativos que dependem de outra camada
60%
Capacidade de reserva (tolerância)
30%

Leituras

Ativos atingidos pela ameaça
7%
Ativos ainda a funcionar
50%
Amplificação do dano inicial
6.8 times
Passos até à liquidação
3
Números, métodos e fontes
PassoCentros de dados a funcionar (%)Redes a funcionar (%)Funcionamento da alimentação (%)
08394100
1727894
2616750
3616722

Métodos

  1. Redes interdependentes Buldyrev and others (2010), Catastrophic cascade of failures in interdependent networks, Nature 464
  2. Cascatas de sobrecarga Motter and Lai (2002), Physical Review E 66
  3. Carga de intermediação Brandes (2001), Journal of Mathematical Sociology 25
  4. Redes multicamada Kivela and others (2014), Journal of Complex Networks 2
  5. Interdependências de infraestruturas Rinaldi, Peerenboom and Kelly (2001), IEEE Control Systems 21(6)
  6. Leque de incerteza (Monte Carlo de parâmetros) Spiegelhalter, Pearson e Short (2011), Visualising uncertainty about the future, Science 333
  7. Sensibilidade de um fator de cada vez (tornado) Saltelli and others (2008), Global Sensitivity Analysis: The Primer, Wiley

Dados

The model runs on the teaching values set by the controls. To calibrate it for a place, read it against these admitted layers and series, which this page’s map can draw:

  • Rodovias principais Natural Earth
  • Ferrovias Natural Earth
  • Portos marítimos Natural Earth
  • Aeroportos Natural Earth
  • Extensões urbanas Natural Earth
  • Luzes noturnas esta noite, falhas (banda dia/noite VIIRS) NASA/NOAA VIIRS (Suomi NPP) · registo nasa.gibs.context-live-layers
  • Eventos naturais (NASA EONET) NASA EONET · registo nasa.eonet.events
Uma cascata por redes interdependentes. Modelo didático ilustrativo: funciona com os valores definidos pelos controles, não com dados medidos. Métodos: redes interdependentes (Buldyrev e outros (2010), Catastrophic cascade of failures in interdependent networks, Nature 464); cascatas de sobrecarga (Motter e Lai (2002), Physical Review E 66); carga de intermediação (Brandes (2001), Journal of Mathematical Sociology 25).
04Em todos os domínios

AI changes cyber risk in every domain’s control systems.

05Métodos e evidência

How AI cyber security is modelled, measured and tested.

O modelo, os seus pressupostos, os seus dados e a sua incerteza, na forma que um cientista ou engenheiro verificaria. As estruturas seguem métodos publicados. Os valores representados são ilustrativos, não conclusões.

Modelo de sistema

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

++−+Inteligência artificialDadosComunidadesEconomiaR de reforçoB equilíbrio
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.
Princípios orientadores FAIR · GO FAIR e FORCE11W3C PROV · World Wide Web ConsortiumOrientação sobre o tratamento consistente das incertezas · IPCCCEOS Analysis Ready Data · Committee on Earth Observation Satellites
06Indicadores

O que é medido e com que qualidade.

Cada indicador leva sua unidade, sua periodicidade, o tipo de fonte por trás dele e a etiqueta de evidência que um valor levaria. Nenhum valor é mostrado: um programa os fornece sob sua própria autoridade.

IndicadorUnidadeCadênciaTipo de fonteRótulo
Vulnerabilities reported in AI frameworks and librariescount of CVE entriesmensalVulnerability databasesREPORTADO
Adversarial robustnessaccuracy under a stated attackper evaluationRobustness testingOBSERVADO
Models with signed provenancepercent of deployed modelsper inventoryModel registriesDECLARADO
AI-related security incidentscount by techniqueanualIncident reportsREPORTADO
Time to detect AI system compromisehoursper incidentIncident response recordsOBSERVADO
Organisations with AI security policiespercent of surveyed organisationsper surveyIndustry surveysREPORTADO
07Fontes

De onde vêm as evidências.

Conjuntos de dados públicos nomeados e registros institucionais, cada um com sua classe de licença. As fontes obtidas por acordo só são usadas nos seus termos e nunca são republicadas.

FonteO que ofereceClasse de licença
MITRE ATLASAdversarial tactics and techniques against AI sistemas, with case studiesPublic, licence to check
NIST adversarial machine learning taxonomyTerminology and taxonomy of attacks and mitigationsPublic domain, US government
National Vulnerability Database and CVE recordsVulnerabilities in AI softwarePublic domain, US government
ENISA reports on AI cybersecurityThreat landscape and good practice for securing AIPublic, ENISA terms
Joint guidance from national cyber agencies on secure AI system developmentSecure design, development, deployment and operationPublic, terms per agency
08Decisões

Quem decide, e a que servem as evidências.

Estúdio de Inteligência artificial e sistemas prepares evidence. The decisions below are taken by the authorities named, under their own rules, and never by the platform.

Quem decideA perguntaEvidências a que esta página serve
Chief information security officersHow to secure AI systems the organisation builds or buysThreat models; supply chain checklists
National cybersecurity agenciesWhat guidance to issue on AI securityIncident and vulnerability evidence
DesenvolvedoresWhich defences to build into models and applicationsRobustness testing; attack taxonomies

Produtos que este tema gera

  • Threat models for AI deployments
  • AI supply chain security checklists
  • Robustness testing protocols
  • Evidence notes on AI-enabled attack trends
09Questões centrais

What security researchers ask.

  • How can inputs be crafted to make a model misbehave, and how robust are defences?
  • Can an attacker poison training data or tamper with model weights in the supply chain?
  • Can private training data or a model itself be extracted through its outputs?
  • How does AI change attackers’ and defenders’ capabilities?
10Mecanismos

The AI attack surface.

Adversarial examples are inputs altered in ways imperceptible to people that cause models to err. Data poisoning inserts malicious examples into training data to plant backdoors. Membership inference and extraction attacks recover information about training data or replicate a model through repeated queries. For language models, jailbreaks and prompt injection bypass intended restrictions.

On the other side, AI can help attackers find vulnerabilities, write malware variants and craft convincing phishing at scale, and help defenders triage alerts, analyse code and detect anomalies. The balance between the two is an open research question.

11Métodos e dados

Security engineering for AI.

  1. 01

    Red-teaming

    Structured attempts to break a system, combining cibersegurança e criptografia and artificial intelligence and machine learning expertise. Red-teaming finds problems but cannot prove their absence.

  2. 02

    Threat modelling

    Mapping assets, adversaries and attack paths across data, training, deployment and supply chain.

  3. 03

    Robustness testing

    Measuring performance under adversarial perturbation. Robustness to known attacks does not imply robustness to new ones.

  4. 04

    Supply-chain integrity

    Provenance of datasets, models and dependencies, with signing and verification. Provenance tooling is still maturing.

13Where this page’s references

Where this page’s references lead.

Os organismos e documentos à luz dos quais esta página é lida. O estado e as edições mudam; cada ligação leva ao organismo emissor.

14Perguntas

Perguntas e respostas.

What is data poisoning?
Manipulating training data so that a model learns hidden behaviours or performs worse, for example responding to a trigger in a way the attacker chooses.
What is MITRE ATLAS?
A knowledge base of adversary tactics and techniques against AI systems, modelled on MITRE ATT&CK, with real-world case studies.
Does AI make cyberattacks easier?
It can lower the effort for some tasks, such as writing phishing messages, according to agency assessments. How much it changes overall attack capability is still being studied.
Does the platform publish attack techniques?
Only at a level that supports defence, following responsible disclosure. Details that would mainly enable harm are not published.
How do I report a vulnerability?
Escrever para o formulário de relato de segurança. A confirmação de recebimento traz uma referência e indica a próxima etapa.
Is open release of model weights a security risk?
It is a trade-off. Open weights enable scrutiny, research and local control, and also make misuse and the removal of safeguards easier. The balance depends on the model’s capabilities.
Can AI defend against AI-enabled attacks?
It can help, but it is not a complete answer. Basic security practices, monitoring and human expertise remain essential.
RelatóriosContribuir

Contribua com evidências a partir desta página.

Relatórios 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.

GCRIInteligência artificial e sistemas autônomosTecnologia exponencialEvidências de Labs
Inteligência artificial e sistemas autônomos

Traga um método, um conjunto de dados ou uma pergunta.

A informação e os pedidos estão abertos hoje. Um pedido recebe uma referência e indica o próximo passo; candidaturas, pagamentos, alocações de computação e registros só abrem quando existem a equipa, os termos e a capacidade de revisão.