MIT AI Risk Repository

Browse AI risks

72 risk entries extracted from 74 frameworks, coded by domain, subdomain, causal entity, intent and timing. Filter, then export the current selection with its licence and citation attached.

Reset Also filtered by framework Hendrycks2023 ×

72 entries · page 1 of 2

  1. "empowering malicious actors to cause widespread harm"

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  2. 22.01.03 · Risk Sub-Category

    Malicious Use (Intentional)

    Persuasive AIs

    "The deliberate propagation of disinformation is already a serious issue, reducing our shared understanding of reality and polarizing opinions. AIs could be used to severely exacerbate this problem by generating personalized disinformation on a larger scale than before. Additionally, as AIs become better at predicting and nudging our behavior, they will become more capable at manipulating us"

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  3. 22.01.01 · Risk Sub-Category

    Malicious Use (Intentional)

    Bioterrorism

    "AIs with knowledge of bioengineering could facilitate the creation of novel bioweapons and lower barriers to obtaining such agents."

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  4. 22.01.02 · Risk Sub-Category

    Malicious Use (Intentional)

    Unleashing AI Agents

    "people could build AIs that pursue dangerous goals’"

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  5. 22.01.04 · Risk Sub-Category

    Malicious Use (Intentional)

    Concentration of Power

    "Governments might pursue intense surveillance and seek to keep AIs in the hands of a trusted minority. This reaction, however, could easily become an overcorrection, paving the way for an entrenched totalitarian regime that would be locked in by the power and capacity of AIs"

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  6. "The immense potential of AIs has created competitive pressures among global players contending for power and influence. This “AI race” is driven by nations and corporations who feel they must rapidly build and deploy AIs to secure their positions and survive."

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  7. 22.02.01 · Risk Sub-Category

    AI Race (Environmental/Structural)

    Military AI Arms Race

    "The development of AIs for military applications is swiftly paving the way for a new era in military technology, with potential consequences rivaling those of gunpowder and nuclear arms in what has been described as the “third revolution in warfare.”

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  8. 22.02.02 · Risk Sub-Category

    AI Race (Environmental/Structural)

    Corporate AI Race

    "Although competition between companies can be beneficial, creating more useful products for consumers, there are also pitfalls. First, the benefits of economic activity may be unevenly distributed, incentivizing those who benefit most from it to disregard the harms to others. Second, under intense market competition, businesses tend to focus much more on short-term gains than on long-term outcomes. With this mindset, companies often pursue something that can make a lot of profit in the short term, even if it poses a societal risk in the long term."

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  9. 22.03.01 · Risk Sub-Category

    Organizational Risks (Accidental)

    Accidents Are Hard to Avoid

    accidents can cascade into catastrophes, can be caused by sudden unpredictable developments and it can take years to find severe flaws and risks (not a quote)

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  10. 22.04.00 · Risk Category

    Rogue AIs (Internal)

    "speculative technical mechanisms that might lead to rogue AIs and how a loss of control could bring about catastrophe"

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  11. 22.04.01 · Risk Sub-Category

    Rogue AIs (Internal)

    Proxy Gaming

    "One way we might lose control of an AI agent’s actions is if it engages in behavior known as “proxy gaming.” It is often difficult to specify and measure the exact goal that we want a system to pursue. Instead, we give the system an approximate—“proxy”—goal that is more measurable and seems likely to correlate with the intended goal. However, AI systems often find loopholes by which they can easily achieve the proxy goal, but completely fail to achieve the ideal goal. If an AI “games” its proxy goal in a way that does not reflect our values, then we might not be able to reliably steer its beh

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  12. 22.04.02 · Risk Sub-Category

    Rogue AIs (Internal)

    Goal Drift

    "Even if we successfully control early AIs and direct them to promote human values, future AIs could end up with different goals that humans would not endorse. This process, termed “goal drift,” can be hard to predict or control. This section is most cutting-edge and the most speculative, and in it we will discuss how goals shift in various agents and groups and explore the possibility of this phenomenon occurring in AIs. We will also examine a mechanism that could lead to unexpected goal drift, called intrinsification, and discuss how goal drift in AIs could be catastrophic."

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  13. 22.04.03 · Risk Sub-Category

    Rogue AIs (Internal)

    Power Seeking

    "even if an agent started working to achieve an unintended goal, this would not necessarily be a problem, as long as we had enough power to prevent any harmful actions it wanted to attempt. Therefore, another important way in which we might lose control of AIs is if they start trying to obtain more power, potentially transcending our own."

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  14. 22.04.04 · Risk Sub-Category

    Rogue AIs (Internal)

    Deception

    "it is plausible that AIs could learn to deceive us. They might, for example, pretend to be acting as we want them to, but then take a “treacherous turn” when we stop monitoring them, or when they have enough power to evade our attempts to interfere with them. "

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  15. 22.01.01.a · Additional evidence

    Malicious Use (Intentional)

    Bioterrorism

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  16. 22.01.01.b · Additional evidence

    Malicious Use (Intentional)

    Bioterrorism

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  17. 22.01.01.c · Additional evidence

    Malicious Use (Intentional)

    Bioterrorism

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  18. 22.01.01.d · Additional evidence

    Malicious Use (Intentional)

    Bioterrorism

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  19. 22.01.02.a · Additional evidence

    Malicious Use (Intentional)

    Unleashing AI Agents

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  20. 22.01.03.a · Additional evidence

    Malicious Use (Intentional)

    Persuasive AIs

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  21. 22.01.03.b · Additional evidence

    Malicious Use (Intentional)

    Persuasive AIs

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  22. 22.01.04.a · Additional evidence

    Malicious Use (Intentional)

    Concentration of Power

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  23. 22.01.04.b · Additional evidence

    Malicious Use (Intentional)

    Concentration of Power

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  24. 22.01.04.c · Additional evidence

    Malicious Use (Intentional)

    Concentration of Power

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  25. 22.02.01.a · Additional evidence

    AI Race (Environmental/Structural)

    Military AI Arms Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  26. 22.02.01.b · Additional evidence

    AI Race (Environmental/Structural)

    Military AI Arms Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  27. 22.02.01.c · Additional evidence

    AI Race (Environmental/Structural)

    Military AI Arms Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  28. 22.02.01.d · Additional evidence

    AI Race (Environmental/Structural)

    Military AI Arms Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  29. 22.02.01.e · Additional evidence

    AI Race (Environmental/Structural)

    Military AI Arms Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  30. 22.02.01.f · Additional evidence

    AI Race (Environmental/Structural)

    Military AI Arms Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  31. 22.02.01.g · Additional evidence

    AI Race (Environmental/Structural)

    Military AI Arms Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  32. 22.02.01.h · Additional evidence

    AI Race (Environmental/Structural)

    Military AI Arms Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  33. 22.02.01.i · Additional evidence

    AI Race (Environmental/Structural)

    Military AI Arms Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  34. 22.02.01.j · Additional evidence

    AI Race (Environmental/Structural)

    Military AI Arms Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  35. 22.02.01.k · Additional evidence

    AI Race (Environmental/Structural)

    Military AI Arms Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  36. 22.02.02.a · Additional evidence

    AI Race (Environmental/Structural)

    Corporate AI Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  37. 22.02.02.b · Additional evidence

    AI Race (Environmental/Structural)

    Corporate AI Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  38. 22.02.02.c · Additional evidence

    AI Race (Environmental/Structural)

    Corporate AI Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  39. 22.02.02.d · Additional evidence

    AI Race (Environmental/Structural)

    Corporate AI Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  40. 22.02.02.e · Additional evidence

    AI Race (Environmental/Structural)

    Corporate AI Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  41. 22.02.02.f · Additional evidence

    AI Race (Environmental/Structural)

    Corporate AI Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  42. 22.02.02.g · Additional evidence

    AI Race (Environmental/Structural)

    Corporate AI Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  43. 22.02.02.h · Additional evidence

    AI Race (Environmental/Structural)

    Corporate AI Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  44. 22.02.02.i · Additional evidence

    AI Race (Environmental/Structural)

    Corporate AI Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  45. 22.02.02.j · Additional evidence

    AI Race (Environmental/Structural)

    Corporate AI Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  46. 22.02.02.k · Additional evidence

    AI Race (Environmental/Structural)

    Corporate AI Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  47. 22.02.02.l · Additional evidence

    AI Race (Environmental/Structural)

    Corporate AI Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  48. 22.02.02.m · Additional evidence

    AI Race (Environmental/Structural)

    Corporate AI Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  49. 22.02.02.n · Additional evidence

    AI Race (Environmental/Structural)

    Corporate AI Race

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

  50. "An essential factor in preventing accidents and maintaining low levels of risk lies in the organizations responsible for these technologies."

    From An Overview of Catastrophic AI Risks (Hendrycks2023)

Informational only, not legal advice. Verify every claim against the linked official sources and consult qualified counsel before acting.