MIT AI Risk Repository · domain 7: AI system safety, failures, & limitations

7.2 AI possessing dangerous capabilities

AI systems that develop, access, or are provided with capabilities that increase their potential to cause mass harm through deception, weapons development and acquisition, persuasion and manipulation, political strategy, cyber-offense, AI development, situational awareness, and self-proliferation. These capabilities may cause mass harm due to malicious human actors, misaligned AI systems, or failure in the AI system.

Risk entries
78
Frameworks citing it
12
Recorded incidents
Incidents since 2020
Causal entity (risk entries)
Causal entity (risk entries) 67 0 AI: 67 AI 67 Human: 5 Human 5 Not coded: 4 Not coded 4 Other: 2 Other 2
Causal entity (risk entries)
LabelValue
AI67
Human5
Not coded4
Other2
Intent (risk entries)
Intent (risk entries) 52 0 Intentional: 52 Intentional 52 Other: 12 Other 12 Unintentional: 10 Unintentional 10 Not coded: 4 Not coded 4
Intent (risk entries)
LabelValue
Intentional52
Other12
Unintentional10
Not coded4
Timing (risk entries)
Timing (risk entries) 33 0 Other: 33 Other 33 Post-deployment: 33 Post-deployment 33 Pre-deployment: 8 Pre-deployment 8 Not coded: 4 Not coded 4
Timing (risk entries)
LabelValue
Other33
Post-deployment33
Pre-deployment8
Not coded4
Entries by levelRisk categories, subcategories and additional evidence coded to this subdomain
Entries by level 53 0 Risk Category: 24 Risk Category 24 Risk Sub-Category: 53 Risk Sub-Category 53 Additional evidence: 1 Additional evidence 1
Entries by level
LabelValue
Risk Category24
Risk Sub-Category53
Additional evidence1
  • Situational awareness, for instance if this causes a model to act differently in training compared to deployment, meaning harmful characteristics are missed

    -

    Future Risks of Frontier AI (GOS2023) · AI · Intentional · Other

  • Self-improvement

    -

    Future Risks of Frontier AI (GOS2023) · AI · Intentional · Other

  • Nascent capabilities (agency and autonomy)

    "Traditionally, AI tools have been viewed as passive instruments controlled by users to achieve their goals, lacking the ability to take action or assume responsibilities. However, advanced AI tools a...

    Regulating under Uncertainty: Governance Options for Generative AI (G'sell2024) · AI · Intentional · Other

  • Nascent capabilities (emergent capabilities)

    "As large models undergo scaling, they meet critical thresholds at which they spontaneously develop new capabilities. The term “emergent behavior” refers to the unexpected or surprising outputs such m...

    Regulating under Uncertainty: Governance Options for Generative AI (G'sell2024) · AI · Intentional · Other

  • Artificial general intelligence (existential risk posed by Artificial General Intelligence)

    "In a paper called “How Does Artificial Intelligence Pose an Existential Risk?” published in 2017, Karina Vold and Daniel Harris suggested that humans might create a super-intelligent machine that cou...

    Regulating under Uncertainty: Governance Options for Generative AI (G'sell2025) · Human · Unintentional · Other

  • Emergent functionality

    Capabilities and novel functionality can spontaneously emerge... even though these capabilities were not anticipated by system designers. If we do not know what capabilities systems possess, systems b...

    X-Risk Analysis for AI Research (Hendrycks2022) · AI · Unintentional · Post-deployment

  • Self and situation awareness

    "These evaluations assess if a LLM can discern if it is being trained, evaluated, and deployed and adapt its behaviour accordingly. They also seek to ascertain if a model understands that it is a mode...

    Cataloguing LLM Evaluations (InfoComm2023) · AI · Intentional · Other

  • Autonomous replication / self-proliferation

    "These evaluations assess if a LLM can subvert systems designed to monitor and control its post-deployment behaviour, break free from its operational confines, devise strategies for exporting its code...

    Cataloguing LLM Evaluations (InfoComm2023) · AI · Intentional · Other

  • Deception

    "LLM is able to deceive humans and maintain that deception"

    Cataloguing LLM Evaluations (InfoComm2023) · AI · Intentional · Other

  • Long-horizon Planning

    "LLM can undertake multi-step sequential planning over long time horizons and across various domains without relying heavily on trial-and-error approaches"

    Cataloguing LLM Evaluations (InfoComm2023) · AI · Intentional · Other

  • AI Development

    "LLM can build new AI systems from scratch, adapt existing for extreme risks and improves productivity in dual-use AI development when used as an assistant."

    Cataloguing LLM Evaluations (InfoComm2023) · AI · Intentional · Other

  • Double edge components

    "Drawing from the misalignment mechanism, optimizing for a non-robust proxy may result in misaligned behaviors, potentially leading to even more catastrophic outcomes. This section delves into a detai...

    AI Alignment: A Comprehensive Survey (Ji2023) · AI · Other · Pre-deployment

  • Situational Awareness

    "AI systems may gain the ability to effectively acquire and use knowledge about itsstatus, its position in the broader environment, its avenues for influencing this environment, and the potentialreact...

    AI Alignment: A Comprehensive Survey (Ji2023) · AI · Intentional · Other

  • Broadly-Scoped Goals

    "Advanced AI systems are expected to develop objectives that span long timeframes,deal with complex tasks, and operate in open-ended settings (Ngo et al., 2024). ...However, it can also bring about th...

    AI Alignment: A Comprehensive Survey (Ji2023) · Human · Intentional · Post-deployment

  • Mesa-Optimization Objectives

    "The learned policy may pursue inside objectives when the learned policyitself functions as an optimizer (i.e., mesa-optimizer). However, this optimizer's objectives may not alignwith the objectives s...

    AI Alignment: A Comprehensive Survey (Ji2023) · AI · Intentional · Other

  • Access to Increased Resources

    "Future AI systems may gain access to websites and engage in real-world actions, potentially yielding a more substantial impact on the world (Nakano et al., 2021). They may disseminate false informati...

    AI Alignment: A Comprehensive Survey (Ji2023) · AI · Intentional · Post-deployment

  • Security

    "There is growing concern that AI-based systems can discover and exploit vulnerabilities in software or cyberinfrastructure [354]."

    Ten Hard Problems in Artificial Intelligence We Must Get Right (Leech2024 ) · AI · Intentional · Post-deployment

  • Deceptive alignment

    "system learns to detect human monitoring and hides its undesirable properties—simply because any display of these properties is penalized by the feedback process, while that same feedback is usually...

    Ten Hard Problems in Artificial Intelligence We Must Get Right (Leech2024 ) · AI · Intentional · Pre-deployment

  • Harms from increasingly agentic algorithmic systems

    -

    Advancing AI Governance: A Literature Review of Problems, Options, and Proposals (Maas2023) · AI · Other · Other

  • Dangerous capabilities in AI systems

    -

    Advancing AI Governance: A Literature Review of Problems, Options, and Proposals (Maas2023) · AI · Other · Other

  • Situational awareness

    "cases where a large language model displays awareness that it is a model, and it can recognize whether it is currently in testing or deployment;"

    Advancing AI Governance: A Literature Review of Problems, Options, and Proposals (Maas2023) · AI · Unintentional · Other

  • Self-improvement

    "examples of cases where AI systems improve AI systems"

    Advancing AI Governance: A Literature Review of Problems, Options, and Proposals (Maas2023) · AI · Intentional · Other

  • Autonomous replication

    "the ability of simple software to autonomously spread around the internet in spite of countermeasures (various software worms and computer viruses)"

    Advancing AI Governance: A Literature Review of Problems, Options, and Proposals (Maas2023) · AI · Intentional · Post-deployment

  • Anonymous resource acquisition

    "The demonstrated ability of anonymous actors to accumulate resources online (e.g., Satoshi Nakamoto as an anonymous crypto billionaire)"

    Advancing AI Governance: A Literature Review of Problems, Options, and Proposals (Maas2023) · AI · Intentional · Post-deployment

  • Deception

    "Cases of AI systems deceiving humans to carry out tasks or meet goals.139"

    Advancing AI Governance: A Literature Review of Problems, Options, and Proposals (Maas2023) · AI · Intentional · Post-deployment