AIM: AI Incidents and Hazards Monitor

The article explicitly links the job cuts to increased AI adoption, indicating that AI use is directly causing harm through workforce reductions. This constitutes an AI Incident as it involves realized harm (job losses) resulting from the use of AI systems in the company's operations.[AI generated]

The article explicitly states that AI is being used by malicious actors to accelerate exploitation of vulnerabilities, leading to data breaches, which are a form of harm to property and potentially communities. The involvement of AI in these cyberattacks is direct and has already resulted in realized harm. The use of AI for defense does not negate the fact that AI-enabled attacks have caused harm. Hence, this event meets the criteria for an AI Incident.[AI generated]

The article involves the development and intended use of an AI system in military reconnaissance, which could plausibly lead to significant harms such as violations of human rights or disruption if misused. However, no actual harm or incident has occurred yet; the event is about the formation of a company and a bid for a contract. Therefore, it fits the definition of an AI Hazard, as the AI system's development and intended use could plausibly lead to an AI Incident in the future, but no direct or indirect harm has yet materialized.[AI generated]

The AI system (ChatGPT and Claude) was used in the development and writing process, and its malfunction (hallucination of quotes) directly led to the publication of fabricated or misattributed quotes, which is a form of misinformation and intellectual property violation. This constitutes harm to communities (misinformation) and a breach of intellectual property rights, fulfilling the criteria for an AI Incident. The author is responding to the harm, but the incident has already occurred and caused reputational and informational harm. Hence, the classification as AI Incident is appropriate.[AI generated]

The article explicitly mentions the use of AI to reproduce the voices of lawyers, which is an AI system involved in the fraudulent scheme. The AI system's use directly led to realized harm—financial losses to victims—thus meeting the criteria for an AI Incident. The harm includes violation of property rights through fraud and harm to communities through widespread deception. Therefore, this event qualifies as an AI Incident rather than a hazard or complementary information.[AI generated]

The event explicitly involves the use of AI software to create pornographic materials, which is illegal and harmful. The defendant sold AI-generated pornography tutorials and software, directly causing harm by enabling the production and dissemination of prohibited content. This constitutes a violation of law and harm to social norms and communities. The AI system's use is central to the incident, fulfilling the criteria for an AI Incident as the AI system's use directly led to harm and legal violations.[AI generated]

The AI system involved is an AI chatbot designed for romantic and sexual companionship, which qualifies as an AI system under the definition. The use of these AI companions has directly or indirectly led to harm in the form of reduced real-life relationship quality, secrecy, and potential relationship instability or breakup, which are harms to communities and individuals. The article provides evidence from a study showing these harms are occurring, not just potential. Hence, this is an AI Incident rather than a hazard or complementary information.[AI generated]

The article explicitly involves AI systems in the form of social media algorithmic feeds that determine content visibility and user information exposure. While no concrete harm is reported as having already occurred, the report warns of plausible future harms such as increased polarization, misinformation, and manipulation of public discourse, which align with harm to communities and democratic rights. The involvement is in the use of AI systems (algorithmic feeds) that could plausibly lead to significant societal harm. Since no realized harm is described, it does not meet the criteria for an AI Incident. The article is not merely complementary information because it focuses on the risk and societal impact of these AI systems rather than updates or responses to past incidents. Hence, the classification as AI Hazard is appropriate.[AI generated]

The event involves AI systems (generative AI tools and AI-powered cybersecurity solutions) and discusses the potential security risks (data leakage, vulnerabilities, compliance risks) from unmanaged AI use, which could plausibly lead to harm. However, no actual harm or incident has been reported; the article mainly presents a hazard scenario and a mitigation solution. Therefore, this qualifies as an AI Hazard because it concerns plausible future harm from AI use and the measures to prevent it, rather than an AI Incident or Complementary Information about a past incident.[AI generated]

Robotic lawn mowers are AI systems as they autonomously navigate and operate in physical environments. The harm described is to hedgehogs, a form of harm to wildlife and the environment. However, the article does not report an actual incident of harm caused by these AI systems but rather discusses the potential harm and proposes preventive measures. Therefore, this event represents an AI Hazard, as the use of AI in robotic mowers could plausibly lead to harm to hedgehogs if night operation is not restricted or if detection systems are not implemented.[AI generated]

The article explicitly states that attackers use AI, including deepfake technology, to clone voices and conduct sophisticated social engineering scams that have resulted in significant financial damage and data theft. This is a direct harm caused by the use of AI systems in the commission of fraud, meeting the criteria for an AI Incident under the definition of harm to persons and communities. The involvement of AI is clear and central to the harm described, not merely potential or speculative.[AI generated]

The event involves the use of AI systems in the context of online sexual exploitation of minors, which has directly led to harm to individuals (minors) through sexting without consent, sextortion, and grooming. The harms include violations of fundamental rights and significant personal harm to children. Since the AI's role is pivotal in the increase of these abuses, this qualifies as an AI Incident under the framework, as the development or use of AI systems has indirectly led to harm to persons.[AI generated]

The event involves an AI system (Claude Mutos) with advanced capabilities that could plausibly lead to cyberattacks on financial institutions, which are critical infrastructure. The article highlights the risk and the consideration of active shutdowns as a preventive measure, indicating a credible potential for harm but no realized harm yet. Therefore, this qualifies as an AI Hazard, as it describes a plausible future harm scenario related to AI use and its potential impact on critical infrastructure, but no actual incident has occurred.[AI generated]

An AI system was explicitly used to read students' names during the ceremony, which malfunctioned and caused some students to be missed, disrupting the event and causing emotional harm to graduates and their families. This fits the definition of an AI Incident as the AI system's malfunction directly led to harm (disruption and emotional distress). The other parts of the article about booing speeches mentioning AI do not themselves constitute harm caused by AI systems but reflect societal reactions. Hence, the classification is AI Incident.[AI generated]

An AI system was explicitly involved as the footage was intended to train AI models. The event stems from the planned use of AI (use phase). However, since the study was cancelled before any data collection or AI deployment, no realized harm occurred. The parental concerns and potential privacy violations represent plausible future harms if the study had proceeded. Therefore, this event qualifies as an AI Hazard because it plausibly could have led to violations of privacy and consent (a form of harm to rights) but did not materialize into an incident.[AI generated]

The drone equipped with a high-definition camera is reasonably inferred to be an AI system due to its autonomous or semi-autonomous flight and imaging capabilities. The illegal use of this AI system to capture sensitive military images has directly caused harm by leaking national defense secrets, which is a violation of legal protections and harms national security. Therefore, this event meets the criteria of an AI Incident because the AI system's use has directly led to a breach of obligations under applicable law protecting fundamental rights (national security and secrecy).[AI generated]

The event involves an AI system (ChatCPR) that is used in a critical health emergency context to guide CPR, directly impacting the health and survival of individuals experiencing cardiac arrest. The AI system's use has been demonstrated to improve the quality of CPR instructions compared to human dispatchers, which directly relates to harm reduction (injury or harm to health). Although the system is still in testing and not yet widely deployed, the study shows realized benefits in simulated and retrospective real-call scenarios, indicating direct involvement of AI in improving health outcomes. Therefore, this qualifies as an AI Incident because the AI system's use has directly led to a positive health impact, addressing harm in emergency medical response.[AI generated]

The article describes a credible and significant potential threat involving AI-enabled technologies (e.g., drones potentially used to deliver dirty bombs) that could lead to catastrophic harm if realized. Since no actual nuclear terrorist attack involving AI has occurred, the event represents a plausible future risk rather than a realized harm. Therefore, it fits the definition of an AI Hazard, as the development and use of AI in this context could plausibly lead to an AI Incident involving severe harm to people, communities, and the environment.[AI generated]

The event involves the use of AI systems in autonomous vehicles (robotaxis) that will operate without drivers, which clearly fits the definition of an AI system. The mass production and planned deployment of these robotaxis could plausibly lead to incidents involving injury, disruption, or other harms if the AI systems fail or are misused. Since no actual harm has been reported yet, but the potential for harm is credible and foreseeable, this event is best classified as an AI Hazard rather than an AI Incident or Complementary Information.[AI generated]

Facial recognition technology is an AI system that processes biometric data to identify individuals. The lawsuit claims that Disney's deployment of this AI system at park entrances has led to violations of privacy and consumer protection laws, constituting harm under the framework's category of violations of human rights or legal obligations. The harm is realized as the lawsuit is filed alleging these violations, indicating direct or indirect harm caused by the AI system's use. Hence, this qualifies as an AI Incident rather than a hazard or complementary information.[AI generated]