AIM: AI Incidents and Hazards Monitor

The article explicitly mentions AI systems being used to enhance cyberattacks, making them more complex and faster, which could plausibly lead to harm such as disruption of critical infrastructure and harm to communities. The discussion is about potential and ongoing threats rather than a specific realized incident causing harm. Therefore, this qualifies as an AI Hazard because it describes a credible risk of future harm stemming from the use of AI in cyberattacks, but does not report a concrete AI Incident with realized harm.[AI generated]

The digital number plates involve GPS tracking technology integrated into vehicles, which qualifies as an AI system due to its data processing and real-time location tracking capabilities. The government's use of this system for continuous surveillance constitutes a breach of privacy and freedom of association and expression, which are human rights violations. Human Rights Watch's explicit concern and the description of the system's operation confirm that harm to rights is occurring. Hence, this event meets the criteria for an AI Incident due to direct harm caused by the AI system's use.[AI generated]

The event involves the use of an AI system (video surveillance algorithm with facial recognition) by the police. The use is alleged to be illegal and outside the strict legal framework, lacking required data protection impact assessments, which constitutes a breach of obligations under applicable law protecting fundamental rights. This meets the criteria for an AI Incident because the AI system's use has directly led to a violation of legal protections and potentially human rights. The CNIL investigation confirms the seriousness of the issue. Therefore, this is classified as an AI Incident.[AI generated]

The article explicitly states that AI is used to manipulate images and videos in political campaigns to attack opponents and influence voters. This manipulation can mislead the public and distort political discourse, which is a form of harm to communities and a violation of rights. Since the harm is occurring through the dissemination of AI-generated disinformation, this qualifies as an AI Incident under the framework, as the AI system's use has directly led to harm in the form of misinformation and manipulation of public opinion.[AI generated]

The event involves an AI system generating highly realistic images, with the research indicating that this capability can reinforce racial biases and contribute to misinformation or deception online. These outcomes constitute harm to communities and potential violations of rights due to biased AI outputs. Since the harm is realized in the form of social bias and misinformation risks, this qualifies as an AI Incident rather than a mere hazard or complementary information.[AI generated]

The article involves AI systems in the form of active intelligent speed assistance technology that uses AI components such as GPS data, speed limit databases, and cameras to monitor and control vehicle speed. The NTSB's recommendations are motivated by a past incident where speeding and impairment caused multiple fatalities, indicating a direct link between speeding and harm. However, the article does not describe an actual AI system malfunction or misuse causing harm; rather, it discusses the potential of AI systems to prevent such harm in the future. Therefore, this event is best classified as an AI Hazard, as the AI system's use could plausibly lead to preventing future AI incidents related to speeding, but no AI-caused harm has occurred yet.[AI generated]

The article describes a credible potential risk from AI systems (deepfakes and bots) that could plausibly lead to harm (disruption of democratic processes and harm to communities through misinformation) but does not report any realized harm or incident. Therefore, it fits the definition of an AI Hazard, as the AI system's use could plausibly lead to an AI Incident in the future, but no direct or indirect harm has yet occurred according to the article.[AI generated]

The event involves an AI system (an algorithm predicting rehab stay length) whose use directly led to harm: denial of medically justified rehabilitation care to older and disabled patients, which constitutes harm to health and a violation of patients' rights. The algorithm's role was pivotal in causing this harm through enforced adherence to its predictions, making this an AI Incident.[AI generated]

The event involves an AI system (Civitai) that facilitates the generation of nonconsensual sexual images, which constitutes a violation of human rights and causes harm to individuals (harm to health and dignity). The harm is ongoing and directly linked to the AI system's use and development. The platform profits from this activity and has features that encourage further harmful AI model creation. Therefore, this qualifies as an AI Incident due to realized harm caused by the AI system's use and malfunction in content moderation.[AI generated]

Google Maps uses AI algorithms to generate navigation routes. In this case, the AI system's routing led tourists into a violent area where they were attacked, causing injury and death. The AI system's use directly led to harm, fulfilling the criteria for an AI Incident. The removal of the route and addition of alerts are responses to the incident but do not change the classification of the event itself.[AI generated]

The event involves an AI-related system (fake AI chatbot ads impersonating Bard) whose malicious use caused direct harm to users through malware installation and credential theft. This fits the definition of an AI Incident because the development and use of an AI system (or its impersonation) directly led to harm to persons and communities. The legal action and description of harm confirm realized harm rather than potential harm, so it is not merely a hazard or complementary information.[AI generated]

The event involves the use of an AI system (deepfake technology) to create realistic fake videos and voices impersonating a public figure. This AI-generated content is used maliciously to deceive and defraud investors, causing harm to individuals (financial loss) and communities (investor trust). The harm is realized as victims are duped by the scam. Therefore, this qualifies as an AI Incident because the AI system's use directly leads to harm through deception and fraud.[AI generated]

The article discusses the FTC's upcoming challenge aimed at preventing harms from voice cloning AI, which is an AI Hazard because it highlights plausible future harms (fraud, misuse of biometric data) that could result from AI voice cloning technology. No specific incident of harm has been reported, so it is not an AI Incident. The event is about encouraging solutions to mitigate potential risks, fitting the definition of an AI Hazard.[AI generated]

The article explicitly mentions AI integration in the logistical support system of the Sukhoi-75 fighter jet. Although no harm or incident has occurred yet, the deployment of AI in a military weapon system with offensive capabilities plausibly could lead to significant harm, such as injury or disruption. The event concerns the development and planned release of an AI-equipped military system, which fits the definition of an AI Hazard. There is no indication of an actual incident or complementary information about responses or mitigation, so it is not an AI Incident or Complementary Information. It is not unrelated because AI involvement is explicit and relevant.[AI generated]

The article explicitly mentions AI-generated malware as a potential threat to data security and integrity, indicating a credible risk that AI systems could be misused to cause harm. However, it does not report any realized harm or incident caused by such malware. The focus is on the potential risks and the need for responsible AI, which aligns with the definition of an AI Hazard rather than an AI Incident or Complementary Information.[AI generated]

The event explicitly mentions the use of an AI system (ChatGPT) in producing a judicial decision that contained false and erroneous legal content. This misuse of AI has directly led to harm by compromising the accuracy and reliability of a legal ruling, which can be seen as a violation of legal rights and judicial standards. Therefore, this qualifies as an AI Incident due to the direct harm caused by the AI system's involvement in the judicial decision-making process.[AI generated]

The event involves AI systems generating audio content that falsely represents a reputable news source, leading to the dissemination of misinformation. This misinformation harms communities by undermining trust in public information and spreading false narratives about significant topics like the pandemic and war. Since the AI-generated content is actively being spread and causing harm, this qualifies as an AI Incident due to realized harm to communities through disinformation.[AI generated]

The article involves AI systems in the context of their potential to replace human jobs, which could plausibly lead to significant economic and social harm in the future. However, no actual harm or incident has occurred yet, and the article is primarily about future risks and expert opinion rather than a realized event. Therefore, it fits the definition of an AI Hazard, as it highlights credible potential harm from AI deployment in the labor market.[AI generated]

The event involves the use of AI systems (machine learning algorithms for image and voice synthesis) by criminals to perpetrate harms including identity theft, extortion, and harassment, which constitute violations of personal security and rights. Since these harms are occurring and the AI's role is pivotal in enabling these crimes, this qualifies as an AI Incident under the framework. The article focuses on the realized harms and the police's response to them, not just potential risks or general information.[AI generated]

The article mentions AI as one of several evolutionary traps that could lead to negative outcomes, but it does not report any realized harm or incident caused by AI systems. The focus is on potential future risks and the need for collective action to avoid these traps. Therefore, it fits the definition of an AI Hazard, as it plausibly points to future harm from AI misalignment but does not describe an actual AI Incident or complementary information about responses or updates.[AI generated]