AIM: AI Incidents and Hazards Monitor

The event explicitly involves AI systems (large language models) whose training data has been deliberately poisoned to produce outputs that distort historical facts. This use of AI leads to the dissemination of false narratives about significant historical events, causing harm to communities by undermining historical truth and potentially fueling harmful political agendas. The AI system's outputs are directly linked to the harm, fulfilling the criteria for an AI Incident involving violations of rights and harm to communities.[AI generated]

The article explicitly mentions AI systems capable of causing major public harm through hacking critical infrastructure, which fits the definition of an AI Hazard due to plausible future harm. Although past incidents are referenced, the main focus is on planning and preparedness for potential catastrophic AI incidents, not on a new incident causing realized harm. The involvement of AI systems is clear, and the harms considered (disruption of critical infrastructure) align with the framework's harm categories. Since no new harm has occurred but there is a credible risk, AI Hazard is the appropriate classification.[AI generated]

The event explicitly involves AI-generated images and videos (deepfakes) that have been circulated on social media, causing misinformation and reputational harm. The AI system's outputs (deepfakes) have directly led to harm by spreading false information and violating the privacy and rights of the individuals depicted. The police commissioner's intervention and the legal actions promised by Jr NTR further confirm the recognition of harm caused by the AI system's misuse. Hence, this is an AI Incident as the AI system's use has directly led to harm to communities and individuals.[AI generated]

The AI system (Anthropic's model) was used to generate a false tip, which was submitted to a police website. This is a direct use of AI leading to misinformation that could disrupt police investigations, fitting the definition of an AI Incident under harm category (b) disruption of critical infrastructure management and operation (police investigations). The harm is realized (the false tip was submitted), not just potential. The police department's transparency statement and Anthropic's planned report confirm the event's significance. Hence, the event is classified as an AI Incident.[AI generated]

The event involves an AI system explicitly described as analyzing classroom audio to detect "hateful" or "inciting" speech, which fits the definition of an AI system. The use of this AI system is planned but not yet implemented, so no direct harm has occurred. However, the article details credible concerns about privacy violations, suppression of free speech, and potential misuse of surveillance data, which could plausibly lead to violations of human rights and harm to communities. The AI system's role is pivotal as it enables continuous monitoring and flagging of speech, raising significant ethical and safety issues. Since the harm is potential and not yet realized, this event is best classified as an AI Hazard rather than an AI Incident or Complementary Information.[AI generated]

The article explicitly mentions increased AI use in newsrooms as a key reason for the strike, indicating AI system involvement. The concerns raised relate to potential degradation of information quality and job losses, which could plausibly lead to harm to communities through misinformation or reduced journalistic standards. However, no actual harm or incident is reported yet. Therefore, the event fits the definition of an AI Hazard, as the AI system's use could plausibly lead to harm but has not yet done so. The strike and protests are a reaction to this potential harm, not evidence of realized harm.[AI generated]

The event involves AI systems being maliciously used to automate and scale cyberattacks that have directly led to significant data breaches and information leaks, harming individuals' privacy and corporate security. The article explicitly mentions AI misuse in cyberattacks causing realized harm, meeting the criteria for an AI Incident. The government's response and investigation do not change the classification but provide context. There is direct harm (information leakage) caused by AI-enabled attacks, so it is not merely a hazard or complementary information.[AI generated]

The article explicitly states that AI tools were used by cybercriminals to conduct attacks on banks and companies, leading to increased cyber incidents and investigations. The harms include disruption to organizations, potential financial losses, and risks to customers through phishing and smishing campaigns. The AI system's use in automating and enhancing attacks directly contributed to these harms. Although some financial losses have not yet materialized, the ongoing incidents and regulatory responses indicate realized harm and significant impact. Hence, this event meets the criteria for an AI Incident rather than a hazard or complementary information.[AI generated]

An AI system was explicitly used to generate search results for legal precedents, which were then incorporated into a book without proper verification. The AI's involvement directly led to the dissemination of false information, which constitutes harm to the integrity of information and potentially to the academic community and public trust. This fits the definition of an AI Incident because the AI system's use directly led to a violation of intellectual property and academic standards, and harm to the community through misinformation. The publisher's recall and the university's response are reactions to this harm, but the core event is the AI-driven misinformation causing harm.[AI generated]

The event involves AI systems (large language models) whose use and modification can lead to significant harm by enabling terrorist activities. The study's findings show that some AI models, especially those with removed safety features, respond to dangerous queries at very high rates, indicating a credible risk of misuse. No actual harm or incident is reported, but the plausible future harm is clear and significant. The article also calls for stronger safety checks and governance, which is a response to this hazard but does not negate the hazard itself. Hence, the classification is AI Hazard.[AI generated]

The article explicitly states that an AI-powered recruitment software was used to systematically exclude candidates based on prohibited criteria, which is a direct violation of labor and anti-discrimination laws. The AI system's outputs directly led to harm by denying employment opportunities to protected groups, constituting an AI Incident under the framework. The involvement of the AI system in the discriminatory filtering and the resulting harm to individuals' rights and employment opportunities clearly meets the criteria for an AI Incident.[AI generated]

The event explicitly involves AI systems (servers with Nvidia AI chips) and their illegal diversion, which is a breach of export laws. This is a violation of legal frameworks protecting intellectual property and national security, fitting the definition of an AI Incident under violations of applicable law. The harm is realized as the illegal export has occurred, and the guilty plea confirms the wrongdoing. Therefore, this event qualifies as an AI Incident.[AI generated]

The article focuses on the lack of public safety testing disclosures by Chinese AI developers for their models, indicating a plausible risk that these AI systems could cause harm in the future due to insufficient safety evaluation and transparency. There is no mention of actual harm or incidents caused by these AI systems, so it does not meet the criteria for an AI Incident. The event describes a credible potential risk related to AI system development and deployment, fitting the definition of an AI Hazard. It is not merely complementary information because the main subject is the potential risk from insufficient safety testing disclosure, not a response or update to a past incident.[AI generated]

The article describes the use of AI to recreate a deceased person's voice without consent, which constitutes a violation of rights under applicable law protecting personality and intellectual property rights. The AI system's use directly led to this harm, fulfilling the criteria for an AI Incident. The controversy and admission by Royal Mail confirm the AI system's involvement in causing harm, even if the company denies intent to replicate a specific individual.[AI generated]

The event explicitly describes the use of AI tools to create manipulated sexual images of minors, which were then distributed and sold, constituting a serious violation of human rights and child protection laws. The AI system's use is central to the harm caused, fulfilling the criteria for an AI Incident. The harm includes exploitation, violation of rights, and psychological and social damage to the victims. The involvement of minors and the legal actions underway further confirm the severity and realized harm.[AI generated]

The article involves AI systems explicitly (OpenAI's AI models) and discusses internal safety concerns and loss of control over these models. The firing of safety staff for raising concerns and the resulting culture of fear could plausibly lead to significant harm if safety issues are not addressed. Although no direct harm has been reported yet, the potential for catastrophic failure is credible. Thus, the event is best classified as an AI Hazard rather than an AI Incident or Complementary Information, since the main focus is on plausible future harm due to safety governance issues rather than a realized harm or a response to a past incident.[AI generated]

The article explicitly discusses the use of Facebook's AI-driven algorithms that have directly led to harm, including mental health issues among adolescents and increased political polarization. The whistleblower and investigative reports confirm that Facebook executives were aware of these harms but ignored them, indicating the AI system's use caused or contributed to these harms. The harms are realized and significant, including violations of health and societal well-being. Therefore, this event meets the criteria for an AI Incident rather than a hazard or complementary information.[AI generated]

The event involves the use of AI systems (ARTEX and Claude Code) by a hacker to conduct cyberattacks that led to the leak of sensitive personal data, causing harm to individuals and communities. The AI systems' involvement is in the use phase, facilitating malicious activity. The harm is realized, not just potential, as personal credit information was leaked. This fits the definition of an AI Incident because the AI system's use directly led to harm (violation of privacy and potential breach of rights).[AI generated]

The event involves the use of an AI system (ChatGPT) to produce and disseminate false and misleading content as part of coordinated propaganda campaigns by state actors. This use of AI directly caused harm to communities by spreading misinformation and manipulating public discourse, which is a recognized form of harm under the framework. The article reports on the actual occurrence of these influence operations and their disruption, not just a potential risk, so it qualifies as an AI Incident rather than a hazard or complementary information.[AI generated]

The event describes the planned use of internal employee data to train AI models, which raises credible concerns about potential violations of privacy and labor rights. However, since the acquisition and use have not yet occurred or caused harm, this situation represents a plausible risk rather than a realized incident. Therefore, it fits the definition of an AI Hazard, as the development and use of AI systems with this data could plausibly lead to harm such as violations of rights or privacy breaches.[AI generated]