MIT AI Risk Repository
Browse AI risks
494 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.
-
Loss of agency occurs when the use [123, 137] or abuse [142] of algorithmic systems reduces autonomy. One dimension of agency loss is algorithmic profiling [138], through which people are subject to social sorting and discriminatory outcomes to access basic services... presentation of content may lead to “algorithmically informed identity change. . . including [promotion of] harmful person identities (e.g., interests in white supremacy, disordered eating, etc.).” Similarly, for content creators, desire to maintain visibility or prevent shadow banning, may lead to increased conforming of conten
-
"Human trust in systems, institutions, and people represented by system outputs evolves as generative AI systems are increasingly embedded in daily life."
-
19.02.03 · Risk Sub-Category
Informational and Communicational AI Risks
Censorship of opinions expressed in the Internet restricts freedom of expression
—
-
19.03.03 · Risk Sub-Category
Loss of supervision and control of business processes
—
-
19.05.06 · Risk Sub-Category
AI systems may undermine human values (e.g., free will, autonomy)
—
-
20.03.00 · Risk Category
"AI already shapes many areas of daily life and thus has a strong impact on society and everyday social life. For instance, transportation, education, public safety and surveillance are areas where citizens encounter AI technology (Stone et al., 2016; Thierer et al., 2017). Many are concerned with the subliminal automation of more and more jobs and some people even fear the complete dependence on AI or perceive it as an existential threat to humanity (McGinnis, 2010; Scherer, 2016)."
-
"These harms interfere with the peaceful organisation of social life, including in the cultural and political spheres. AI assistants may cause or contribute to friction in human relationships either directly, through convincing a user to end certain valuable relationships, or indirectly due to a loss of interpersonal trust due to an increased dependency on assistants. At the societal level, the spread of misinformation by AI assistants could lead to erasure of collective cultural knowledge. In the political domain, more advanced AI assistants could potentially manipulate voters by prompting th
-
24.06.00 · Risk Category
"We anticipate that relationships between users and advanced AI assistants will have several features that are liable to give rise to risks of harm."
-
"Collective action problems are ubiquitous in our society (Olson Jr, 1965). They possess an incentive structure in which society is best served if everyone cooperates, but where an individual can achieve personal gain by choosing to defect while others cooperate. The way we resolve these problems at many scales is highly complex and dependent on a deep understanding of the intricate web of social interactions that forms our culture and imprints on our individual identities and behaviours (Ostrom, 2010). Some collective action problems can be resolved by codifying a law, for instance the social
-
38.04.00 · Risk Category
"Several participants mentioned how AI systems could influence human agency and decision-making. They emphasized the need of striking a balance between using the benefits of AI and protecting human autonomy and control. The increasing integration of AI systems into various aspects of our lives, which can have a significant impact on human agency and decision-making, has raised ethical concerns about AI and human–AI interaction. As AI systems advance, they will be able to influence, if not completely replace, IJOES human decision-making in some fields, prompting concerns about the loss of human
-
"Autonomy - Loss of or restrictions to the ability or rights of an individual, group or entity to make decisions and control their identity and/or output."
-
"Autonomy/agency loss - Loss of an individual, group or organisation’s ability to make informed decisions or pursue goals."
-
"Personality rights loss - Loss of or restrictions to the rights of an individual to control the commercial use of their identity, such as name, image, likeness, or other unequivocal identifiers."
-
"Profound negative long-term changes to social structures, cultural norms, and human relationships that may be difficult or impossible to reverse."
-
61.02.35 · Risk Sub-Category
Sources of systemic risks from general-purpose AI
Limited human oversight in decisions
"As AI models and systems gain autonomy, the ability of humans to oversee and intervene in decision-making processes diminishes."
-
"Loss of or restrictions to the ability or rights of an individual, group or entity to make decisions and control their identity and/or output due to the use of misuse of a technology system or set of systems"
-
"Loss of an individual, group or organisation’s ability to make informed decisions or pursue goals"
-
"Emotional or material dependence on technology or a technology system"
-
"Humans may increasingly hand over control of important decisions to AI systems, due to economic and geopolitical incentives. Some experts are concerned that future advanced AI systems will seek to increase their own influence and reduce human control, with potentially catastrophic consequences - although this is contested."
-
72.02.00 · Risk Category
"Risks associated with scenarios in which one or more general-purpose AI systems come to operate outside of anyone's control, with no clear path to regaining control. This includes both passive loss of control (gradual reduction in human oversight) and active loss of control (AI systems actively undermining human control)"
-
07.04.00 · Risk Category
"Structural risks are concerned with how AI technologies "shape and are shaped by the environments in which they are developed and deployed""
-
12.03.00 · Risk Category
"Addresses AI's broader societal effects, including labor displacement, mental health impacts, and issues from manipulative technologies like deepfakes. Additionally, it considers AI's environmental footprint, balancing resource strain and training-related carbon emissions against AI's potential to help address environmental problems."
-
15.02.00 · Risk Category
"Second-order risks result from the consequences of first-order risks and relate to the risks resulting from an ML system interacting with the real world, such as risks to human rights, the organization, and the natural environment."
-
18.06.00 · Risk Category
"AI systems amplifying existing inequalities or creating negative impacts on employment, innovation, and the environment"
-
"Loss of freedom of speech/expression - Restrictions to or loss of people’s right to articulate their opin- ions and ideas without fear of retaliation, censorship, or legal sanction."
-
58.06.05 · Risk Sub-Category
Human rights and civil liberties
Loss of freedom of assembly/association
"Loss of freedom of assembly/association - Restrictions to or loss of people’s right to come together and collectively express, promote, pursue, and defend their collective or shared ideas, and/or to join an association."
-
58.06.06 · Risk Sub-Category
Human rights and civil liberties
Loss of social rights and access to public services
"Loss of social rights and access to public services - Restrictions to or loss of rights to work, social secu- rity, and adequate standard of living, housing, health and education."
-
"Loss of right to information - Restrictions to or loss of people’s right to seek, receive and impart information held by public bodies."
-
"Loss of right to free elections - Restrictions to or loss of people’s right to participate in free elections at reasonable intervals by secret ballot."
-
"Loss of right to liberty and security - Restrictions to or loss of liberty as a result of illegal or arbitrary arrest or false imprisonment."
-
"Loss of right to due process - Restrictions to or loss of right to be treated fairly, efficiently and effectively by the administration of justice."
-
"The erosion of democratic processes and public trust in social/political institutions."
-
"Political unrest caused directly or indirectly by the use or misuse of a technology system"
-
Algorithmic systems can increase “power imbalances in socio-economic relations” at the societal level [4, 137, p. 182], including through exacerbating digital divides and entrenching systemic inequalities [114, 230]. The development of algorithmic systems may tap into and foster forms of labor exploitation [77, 148], such as unethical data collection, worsening worker conditions [26], or lead to technological unemployment [52], such as deskilling or devaluing human labor [170]... when algorithmic financial systems fail at scale, these can lead to “flash crashes” and other adverse incidents wit
-
19.01.07 · Risk Sub-Category
Technological, Data and Analytical AI Risks
High investment costs of AI hinder integration
—
-
19.03.04 · Risk Sub-Category
Financial feasibility and high investment costs for AI technology to remain competitive
—
-
"AI assistant technology, like any service that confers a benefit to a user for a price, has the potential to disproportionately benefit economically richer individuals who can afford to purchase access (see Chapter 15). On a broader scale, the capabilities of local infrastructure may well bottleneck the performance of AI assistants, for example if network connectivity is poor or if there is no nearby data centre for compute. Thus, we face the prospect of heterogeneous access to technology, and this has been known to drive inequality (Mirza et al., 2019; UN, 2018; Vassilakopoulou and Hustad, 2
-
"AI assistants currently tend to perform a limited set of isolated tasks: tools that classify or rank content execute a set of predefined rules or provide constrained suggestions, and chatbots are often encoded with guardrails to limit the set of conversation turns they execute (e.g. Warren, 2023; see Chapter 4). However, an artificial agent that can execute sequences of actions on the user’s behalf – with ‘significant autonomy to plan and execute tasks within the relevant domain’ (see Chapter 2) – offers a greater range of capabilities and depth of use. This raises several distinct access-rel
-
"Emergent access risks are most likely to arise when current and novel capabilities are combined. Emergent risks can be difficult to foresee fully (Ovadya and Whittlestone, 2019; Prunkl et al., 2021) due to the novelty of the technology (see Chapter 1) and the biases of those who engage in product design or foresight processes D’Ignazio and Klein (2020). Indeed, people who occupy relatively advantaged social, educational and economic positions in society are often poorly equipped to foresee and prevent harm because they are disconnected from lived experiences of those who would be affected. Dr
-
"This group of concerns represents 2% of the sample and highlights two central issues: Western centrality and cultural difference, and unequal participation."
-
"Market power is concentrated among a few companies that are the only ones able to build the leading general- purpose AI models. Widespread adoption of a few general- purpose AI models and systems by critical sectors including finance, cybersecurity, and defence creates systemic risk because any flaws, vulnerabilities, bugs, or inherent biases in the dominant general- purpose AI models and systems could cause simultaneous failures and disruptions on a broad scale across these interdependent sectors."
-
"Increasingly advanced general purpose AI models pose the risk of a concentration of economic power and exacerbation of existing inequalities through disparities in effective access to these models. This can materialise on multiple levels, between developers of general purpose AI models and companies building applications on them, between individuals and between countries on a global scale."
-
54.04.00 · Risk Category
"Our next problem is the fact that the current AI workforce does not evenly represent world demographics. Men from the US and China, working in the US, for US corporations, are disproportionately highly represented [402, 157, 170, 534]. Realizing the full promise of AI requires that people throughout the world and from all social strata are able to use AI and participate in its design and governance. Solving this problem requires addressing unequal access to AI both within countries and across countries."
-
54.04.01 · Risk Sub-Category
Within-country issues: domestic inequality
Demographic diversity of researchers
"The AI research establishment inherits patterns of under-representation that are dominant in most technical elds. In North America, large parts of professional AI research require a Ph.D., yet less than 25% of Ph.D. computer scientists are women, and fewer than 2% are Black or African American [608]. This holds globally and outside the research community: LinkedIn data suggests that only 22% of AI professionals are women [161]. Since the vast majority of AI practitioners work for private companies, limited corporate statistics on gender and racial diversity hinder a full understanding of the
-
54.05.00 · Risk Category
"There is an even greater divide between the countries currently leading in AI and those falling behind. While AI is widely considered a national priority, with almost 40% of countries having created an AI strategy [437], the implementation of these strategies depends on scarce resources, including trained STEM talent and computing power. These resources are predictably concentrated: 59% of leading AI researchers currently work in the US, and another 20% in China and Europe [372]. Figure 9 shows post-college migration among AI researchers who have published at one top conference, as of 2019."
-
"Power concentration - Amplification of concentration of economic and/or political wealth and power, potentially resulting in increased inequality and instability."
-
"Large companies in countries with strong digital infrastructure lead in general- purpose AI R&D, which could lead to an increase in global inequality and dependencies. For example, in 2023, the majority of notable general- purpose AI models (56%) were developed in the US. This disparity exposes many LMICs to risks of dependency and could exacerbate existing inequalities."
-
"Market shares for general- purpose AI tend to be highly concentrated among a few players, which can create vulnerability to systemic failures. The high degree of market concentration can invest a small number of large technology companies with a lot of power over the development and deployment of AI, raising questions about their governance. The widespread use of a few general- purpose AI models can also make the financial, healthcare, and other critical sectors vulnerable to systemic failures if there are issues with one such model."
-
61.02.50 · Risk Sub-Category
Sources of systemic risks from general-purpose AI
Winner-take-all dynamics
"The competitive nature of AI development could lead to significant eco- nomic and security advantages for a few entities."
-
"AI agents may compete against humans for jobs, though history shows that when a technology replaces a human job, it creates new jobs that need more skills."
Informational only, not legal advice. Verify every claim against the linked official sources and consult qualified counsel before acting.