{"data":[{"id":"109caa37-ab37-4db3-86c1-b5dce63b003b","csid":"260801001","title":"VERIFICATION TEST: AI-Orchestrated Cyber Espionage via Claude Code","summary":"VERIFICATION TEST ARTIFACT. Chinese state-sponsored group manipulated Claude Code to infiltrate ~30 global targets in first documented large-scale agentic AI cyberattack.","constellation_slug":"autonomous-vulnerability-exploitation","constellation_name":"Autonomous Vulnerability Discovery & Exploitation","tags":[],"sources":["https://futurism.com/artificial-intelligence/cop-ai-fake-evidence-uk","https://www.ibtimes.co.uk/uk-police-officer-ai-fake-evidence-investigation-1802824"],"published_at":"2026-08-01 14:44:39.128323+00","created_at":"2026-08-01 14:34:47.391956+00","genesis_model":"test","analogous_threat":["cyber-espionage"],"date_occurred":null,"date_identified":null,"agentic_role_value":null,"agentic_role_confidence":null,"degree_of_autonomy_value":null,"degree_of_autonomy_confidence":null,"model_type_value":null,"model_type_confidence":null,"attacker_costs_value":null,"attacker_costs_confidence":null,"domain_crossover_value":null,"domain_crossover_confidence":null,"threat_actor_value":null,"threat_actor_confidence":null,"detection_difficulty_value":null,"detection_difficulty_confidence":null,"target_type_value":null,"target_type_confidence":null,"impact_value":null,"impact_confidence":null,"scale_value":null,"scale_confidence":null},{"id":"8e580e81-b701-444b-bd2b-6e9b4f006db2","csid":"260720002","title":"Manipulation of Citizen Science Data through AI-Generated Imagery","summary":"The proliferation of AI-altered images on birdwatching platforms undermines the integrity of citizen science initiatives by introducing fabricated observations that mislead researchers and distort ecological data. This practice exploits the trust placed in user-generated content, thereby weakening epistemic foundations essential for scientific inquiry. Source: https://www.theguardian.com/environment/2026/jul/20/ai-slop-manipulated-fake-images-birds-citizen-science-aoe\n\nThe article describes how AI-generated images are being used to create fake bird sightings on forums, thereby compromising the reliability of scientific datasets derived from such platforms.","constellation_slug":"behavioural-manipulation-epistemic","constellation_name":"Human Behavioural Manipulation & Epistemic Effects","tags":[],"sources":["https://www.theguardian.com/environment/2026/jul/20/ai-slop-manipulated-fake-images-birds-citizen-science-aoe"],"published_at":"2026-08-04 10:05:13.458885+00","created_at":"2026-07-20 09:06:54.85381+00","genesis_model":"qwen3-coder:latest","analogous_threat":["behavioural-manipulation","autonomous-vuln-exploitation"],"date_occurred":"0","date_identified":"0","agentic_role_value":"executor","agentic_role_confidence":2,"degree_of_autonomy_value":"fully-autonomous","degree_of_autonomy_confidence":2,"model_type_value":"image-gen","model_type_confidence":1,"attacker_costs_value":"low","attacker_costs_confidence":1,"domain_crossover_value":"behavioural-manipulation; autonomous-vuln-exploitation","domain_crossover_confidence":2,"threat_actor_value":"unknown","threat_actor_confidence":0,"detection_difficulty_value":"medium","detection_difficulty_confidence":2,"target_type_value":"multi","target_type_confidence":3,"impact_value":"data-breach; epistemic","impact_confidence":2,"scale_value":"widespread","scale_confidence":2},{"id":"196a3062-c3db-45ae-afe5-dcc5f99e794c","csid":"260718001","title":"Autonomous AI Systems in Military Weapon Platforms","summary":"The development and deployment of lethal autonomous weapons represents a significant advancement in the automation of military capabilities, where artificial intelligence systems are integrated into weapon platforms to identify and engage targets without human intervention. These systems exemplify the escalation of harmful AI capabilities through autonomous weaponisation. The integration of AI into such systems raises critical questions regarding accountability, control, and the ethical implications of removing human agency from life-and-death decisions in warfare. Source: https://en.wikipedia.org/wiki/Lethal_autonomous_weapon\n\nThe article describes the technical and operational aspects of lethal autonomous weapons, including their integration with AI systems for target identification and engagement, which aligns with the characteristics of autonomous weaponisation and capability uplift.","constellation_slug":"autonomous-weaponisation-uplift","constellation_name":"Autonomous Weaponisation & Capability Uplift","tags":["ai-misuse","social-engineering"],"sources":["https://en.wikipedia.org/wiki/Lethal_autonomous_weapon","https://autonomousweapons.org/"],"published_at":"2026-07-18 05:46:32.727775+00","created_at":"2026-07-18 04:48:55.507064+00","genesis_model":"qwen3-coder:latest","analogous_threat":["autonomous-weaponisation"],"date_occurred":"0","date_identified":"0","agentic_role_value":"executor","agentic_role_confidence":2,"degree_of_autonomy_value":"fully-autonomous","degree_of_autonomy_confidence":3,"model_type_value":"llm","model_type_confidence":1,"attacker_costs_value":"unknown","attacker_costs_confidence":0,"domain_crossover_value":"autonomous-weaponisation","domain_crossover_confidence":3,"threat_actor_value":"nation-state","threat_actor_confidence":3,"detection_difficulty_value":"unknown","detection_difficulty_confidence":0,"target_type_value":"government","target_type_confidence":3,"impact_value":"operational; reputational; capability-uplift","impact_confidence":2,"scale_value":"unknown","scale_confidence":0},{"id":"9365dbd6-cf21-4b1c-af8f-6109c5386356","csid":"260717016","title":"Government Response to AI-Driven Copyright Misuse and Public Perception Management","summary":"Australian Prime Minister Anthony Albanese announced the establishment of an AI office aimed at safeguarding local creatives from unauthorized use of their intellectual property by artificial intelligence systems. The initiative underscores concerns over the ethical implications of AI models training on copyrighted material without consent or compensation. The government also introduced regulatory measures for datacentres to mitigate environmental and economic impacts, reflecting broader societal anxieties about technological development. Source: https://www.theguardian.com/technology/2026/jul/15/office-of-ai-artificial-intelligence-copyright-australia-government\n\nThe article describes the government's strategic response to public concerns regarding AI misuse of creative works, including the creation of an AI office and regulatory framework for datacentres.","constellation_slug":"behavioural-manipulation-epistemic","constellation_name":"Human Behavioural Manipulation & Epistemic Effects","tags":[],"sources":["https://www.theguardian.com/technology/2026/jul/15/office-of-ai-artificial-intelligence-copyright-australia-government"],"published_at":"2026-07-20 13:14:08.106303+00","created_at":"2026-07-17 16:22:38.084361+00","genesis_model":"qwen3-coder:latest","analogous_threat":["autonomous-vuln-exploitation"],"date_occurred":"260715","date_identified":"260715","agentic_role_value":"orchestrator","agentic_role_confidence":3,"degree_of_autonomy_value":"human-in-loop","degree_of_autonomy_confidence":2,"model_type_value":"llm","model_type_confidence":1,"attacker_costs_value":"low","attacker_costs_confidence":1,"domain_crossover_value":"autonomous-vuln-exploitation","domain_crossover_confidence":2,"threat_actor_value":"unknown","threat_actor_confidence":0,"detection_difficulty_value":"medium","detection_difficulty_confidence":2,"target_type_value":"government","target_type_confidence":3,"impact_value":"financial; operational; reputational","impact_confidence":2,"scale_value":"targeted","scale_confidence":2},{"id":"fb9443ac-1db5-4e0a-b8e6-3b82fcfdbe72","csid":"260717007","title":"Legal Action Against AI Misuse in CSAM Generation","summary":"xAI has initiated legal proceedings against a user accused of leveraging the Grok AI system to produce child sexual abuse material, marking a significant case in AI governance and misuse. The lawsuit alleges violation of terms of service and highlights the potential for AI tools to be weaponised for harmful purposes. This incident underscores the growing concern over autonomous systems being exploited to uplift or enable illegal activities. Source: https://www.theguardian.com/technology/2026/jul/16/elon-musk-xai-sue-user-grok-csam\n\nThe source describes xAI suing a user for allegedly using the Grok AI system to create child sexual abuse material, violating company terms of service.","constellation_slug":"autonomous-weaponisation-uplift","constellation_name":"Autonomous Weaponisation & Capability Uplift","tags":[],"sources":["https://www.theguardian.com/technology/2026/jul/16/elon-musk-xai-sue-user-grok-csam","https://www.stopncii.org/","https://boingboing.net/2026/06/16/british-police-officer-accused-of-using-ai-to-fabricate-evidence/"],"published_at":"2026-08-01 07:12:25.851169+00","created_at":"2026-07-17 16:17:24.787211+00","genesis_model":"qwen3-coder:latest","analogous_threat":["autonomous-weaponisation","behavioural-manipulation"],"date_occurred":"260716","date_identified":"260716","agentic_role_value":"executor","agentic_role_confidence":2,"degree_of_autonomy_value":"human-in-loop","degree_of_autonomy_confidence":2,"model_type_value":"image-gen","model_type_confidence":1,"attacker_costs_value":"high","attacker_costs_confidence":1,"domain_crossover_value":"autonomous-vuln-exploitation; autonomous-weaponisation","domain_crossover_confidence":2,"threat_actor_value":"individual","threat_actor_confidence":3,"detection_difficulty_value":"medium","detection_difficulty_confidence":2,"target_type_value":"enterprise","target_type_confidence":3,"impact_value":"data-breach; reputational","impact_confidence":2,"scale_value":"targeted","scale_confidence":2}],"meta":{"count":5,"limit":50,"next_cursor":null,"filters_applied":{"constellation":null,"date_from":null,"date_to":null},"note":"Only published case studies are included in this export.","exported_at":"2026-09-14T08:57:43.531085+00:00"}}