AIM: AI Incidents and Hazards Monitor

The event involves the use of AI systems (smart glasses integrated with large language models) to cheat in exams, which directly harms the integrity of educational assessments and the fairness of academic evaluation, impacting communities and institutions. The article provides concrete examples of detected misuse and institutional countermeasures, indicating realized harm rather than potential harm. Hence, it meets the criteria for an AI Incident rather than a hazard or complementary information.[AI generated]

The viral video is explicitly described as AI-generated and falsely suggests a collaboration that does not exist, indicating the use of AI to create misleading content. While no direct harm such as injury, rights violations, or disruption has occurred, the potential for misinformation and reputational damage is credible. The event fits the definition of an AI Hazard because the AI system's use in generating deepfake content could plausibly lead to harm, such as misleading the public or damaging reputations, even though no such harm has materialized yet according to the article.[AI generated]

The event explicitly involves AI systems used for content moderation and recommendation on Instagram. The AI's malfunction or circumvention by malicious actors has directly led to harm by exposing minors to explicit content and malware risks. This meets the criteria for an AI Incident because the AI system's failure to effectively moderate content has caused realized harm to vulnerable users and breaches legal protections. Therefore, the event is classified as an AI Incident.[AI generated]

The event involves an AI system (Freedom2hear) actively used to moderate social media content and protect women cricketers from online abuse, which is a form of harm to health and wellbeing. The AI system's deployment has directly led to the removal of harmful content and restrictions on offenders, thus preventing or mitigating harm. Since the harm is realized and the AI system's role is pivotal in addressing it, this qualifies as an AI Incident.[AI generated]

The event involves AI systems (AI cameras detecting mobile phone use while driving) and their intended use, but no actual harm or enforcement has occurred. The article focuses on preparatory steps, legal changes, and pilot testing, indicating potential future use but no current incident or harm. Therefore, it represents a plausible future risk scenario but not an incident or complementary information about harm or responses to harm.[AI generated]

The event involves the development of an AI-enabled military drone system, which qualifies as an AI system due to its autonomous capabilities and integration of sensors and weapons. The article focuses on the collaboration and development phase without any realized harm or malfunction. Given the military nature and potential for harm, this constitutes an AI Hazard, as the development and future deployment of such systems could plausibly lead to incidents involving injury, violations of rights, or harm to communities. There is no indication of actual harm yet, so it is not an AI Incident. It is not merely complementary information because the main focus is on the development of a potentially hazardous AI system, not on responses or updates to past incidents.[AI generated]

The software "Tangles" and its module "Webloc" involve AI systems for analyzing online activities and extracting location data. The event involves the use of these AI systems by a government agency, which leads to indirect harm in the form of potential violations of privacy and fundamental rights due to mass data collection and surveillance without effective control. Although no specific incident of harm is described, the event highlights ongoing use and concerns about rights violations and structural surveillance expansion, which constitute realized harm under violations of human rights and fundamental rights. Therefore, this qualifies as an AI Incident.[AI generated]

The scam used AI-generated deepfake technology to create a convincing fraudulent video, which directly caused the victim to invest large sums of money into a fake platform, resulting in substantial financial and emotional harm. The AI system's role was pivotal in enabling the deception and consequent harm. The event clearly meets the criteria for an AI Incident as the AI system's use directly led to realized harm (financial loss and psychological distress).[AI generated]

The article involves an AI system explicitly (Palantir's Maven Smart System) and its potential use in military decision-making, which could plausibly lead to significant harms if misused or malfunctioning, such as harm to national security or human rights. However, since the AI system is only being considered and tested in exercises without any reported harm or malfunction, this constitutes a potential risk rather than an actual incident. The concerns about dependence on foreign AI and information leaks further support the classification as a hazard. Thus, the event fits the definition of an AI Hazard rather than an AI Incident or Complementary Information.[AI generated]

An AI system was used to generate part of a video that misrepresents factual reality, misleading voters about political affiliations. This directly leads to harm to communities by spreading false information and potentially influencing electoral outcomes. Therefore, this qualifies as an AI Incident due to the realized harm caused by the AI-generated misleading content.[AI generated]

The AI system was used to generate fabricated images that led to misinformation and public disruption, which is a form of harm to communities. The event describes realized harm through the spread of false information causing authorities to take action. Therefore, it qualifies as an AI Incident because the AI-generated content directly led to social disruption and misuse of public resources. The event is not merely a potential risk but an actual incident involving AI misuse causing harm.[AI generated]

The event involves an AI system (Amazon Q) that automatically executes commands from project configuration files, which is a use of AI to assist coding. The vulnerability allowed attackers to execute arbitrary code and access sensitive credentials, directly causing harm to property and security. The harm is realized, not just potential, as demonstrated by the proof-of-concept attack. Therefore, this qualifies as an AI Incident because the AI system's malfunction directly led to significant harm.[AI generated]

The article explicitly mentions AI technology being used to create fake images and videos of Mỹ Tâm, which are then disseminated on social media. This misuse of AI has directly harmed the celebrity's reputation and risks misleading the public, which qualifies as harm to communities and a violation of rights related to personal dignity and reputation. The involvement of AI in generating manipulated content that causes real harm fits the definition of an AI Incident rather than a hazard or complementary information.[AI generated]

The presence of AI systems is clear, specifically autonomous AI agents involved in cybersecurity. The article centers on the potential risks these systems introduce, including adaptive behaviors that evade traditional security measures, which could plausibly lead to cybersecurity incidents harming organizations or critical infrastructure. However, no actual incident or harm is described; the discussion is forward-looking and strategic, focusing on risk identification and governance challenges. This aligns with the definition of an AI Hazard, where the development or use of AI systems could plausibly lead to harm but no harm has yet occurred. It is not Complementary Information because it is not updating or responding to a past incident, nor is it Unrelated since it directly concerns AI risks. It is not an AI Incident because no realized harm or violation is reported.[AI generated]

The event involves the use of an AI-enabled software tool that manipulates biometric authentication processes, which is an AI system by definition. The criminal use of this AI system directly caused harm by enabling unauthorized access to bank accounts and facilitating illegal financial transactions, fulfilling the criteria for an AI Incident. The harm is materialized and significant, involving violations of property rights and financial theft. The involvement of AI in the development and use of the biometric spoofing software is central to the incident, not merely incidental or potential. Hence, the classification as AI Incident is appropriate.[AI generated]

The event explicitly involves the use of an AI system (an AI application solving exam questions from photos) during the exam, which is a misuse of AI leading to a legal violation. This misuse caused harm by undermining the integrity of the exam process, which is a breach of legal obligations. Therefore, this qualifies as an AI Incident because the AI system's use directly led to a violation of applicable law and harm to the examination system's fairness.[AI generated]

The article involves AI systems in the form of military drones, which typically rely on AI for navigation, targeting, and operational autonomy. The event concerns the development and planned use of these AI-enabled systems, which could plausibly lead to harm in the future (e.g., in combat scenarios). Since no actual harm or incident has occurred yet, and the article is about planned expansion and training, this constitutes an AI Hazard rather than an AI Incident. It is not merely complementary information because the focus is on the potential for harm through the deployment of AI-enabled drones, not on responses or updates to past incidents.[AI generated]

The event explicitly involves advanced AI chips, which are AI system components, and their illicit movement is being prevented by customs authorities. The chips' smuggling attempt is linked to circumventing export controls designed to prevent unauthorized AI technology transfer, which could plausibly lead to harms such as enabling unauthorized AI development or military applications. Since no actual harm has occurred yet and the event is about interception and investigation, it fits the definition of an AI Hazard. There is no indication of realized harm or incident resulting from the AI system's malfunction or misuse at this stage, so it is not an AI Incident. The event is more than general AI-related news, as it concerns a specific event with potential for harm, so it is not Complementary Information or Unrelated.[AI generated]

The article explicitly mentions that AI and automation mistakes undermined vehicle quality, leading Ford to hire more humans to correct these errors. This shows the AI system's use directly contributed to harm (quality defects) that affected the vehicles, which is harm to property and consumers. Therefore, this qualifies as an AI Incident due to the realized harm caused by the AI system's malfunction or misapplication in production processes.[AI generated]

The article explicitly mentions AI agent advisers making consequential investment decisions, indicating the presence of AI systems. The lawmakers' concerns about investor protection and market integrity imply potential risks of harm, but no actual harm or incident is reported. The focus is on regulatory oversight and the need for guardrails to prevent possible future harms. Hence, this qualifies as an AI Hazard, as the development and use of these AI trading agents could plausibly lead to incidents affecting investors and market stability if left unregulated.[AI generated]