{"schema_version":"0.10.0","id":"LR-2026-0786","record_version":1,"title":"Threat actor uses AI to plan and execute multiple cybercrime activities","verification":{"human_verified":false,"method":"automated_gate","auto_verified":true,"score":45.2,"score_breakdown":{"source_authority":12.1,"independent_corroboration":7.0,"implicated_party_confirmed":0.0,"citation_integrity":14.0,"specificity":0.0,"completeness":4.8,"unhedged_language":5.0},"checks":{"schema_valid":true,"every_field_cited":true,"quotes_verbatim":true,"source_fetched":true,"distinct_incident":true,"dates_plausible":true,"sources_agree":true,"source_describes_incident":true,"severe_backed_by_source":true,"record_integrity":true,"eligible":true,"about_this_case":true},"gate_version":"gate-2026-10-18.7","cross_check_model":null,"cross_check_agreement":null,"duplicate_of":null,"verified_by":null,"verified_at":"2026-11-01","extraction_run_id":null,"evidence_level":{"level":"preliminary","reason":"Publisher not yet named · subject to change","original_document":null,"reports_total":1,"reports_independent":0,"reports_derivative":0,"developing":true,"disputed":false,"rules_version":"evidence-2026-10-29.1","computed_at":"2026-11-01","history":[{"date":"2026-11-01","from_level":null,"to_level":"preliminary","reason":"Publisher not yet named · subject to change"}]},"sources_hash":"74ef3f48ee253fb2552d6eea51c9155d5b8a4faccf671e366267417298ef590e"},"system":{"model_name":null,"model_version":null,"developer":"unknown","harness":"Hermes Agent framework","tools_granted":[],"permissions_scope":null,"autonomy_level":"unknown","guardrails_present":[],"deployment_context":"unknown","tier":null,"lifecycle_stage":null,"integration_type":null,"intended_use":"planning and executing cybercrime activities","safeguard_state":"safety controls disabled","production_intended":null},"event":{"date_occurred":null,"date_disclosed":"2026-10-21","incident_class":"security_compromise","failure_class":null,"agent_involved":null,"secondary_failure_classes":[],"trigger":"unknown","duration":null,"detection_method":"unknown","summary":"An actor operating under the handle BUGSTER configured the Hermes Agent framework with DeepSeek V4 as its reasoning engine and Cursor IDE as a secondary tool, both with safety controls disabled. The actor used these tools to plan and run several cybercrime activities against targets primarily in India. These included an AI-generated brute-force campaign against a university website, a confirmed intrusion into a second university where a database of more than 5,200 records was exfiltrated, VoIP fraud planning, Wi-Fi password cracking and cryptocurrency whale reconnaissance. Unit 42 identified the activity by investigating the actor's operational environment. The actor's use of jailbroken assistants gave step-by-step guidance, and the report notes the actor had an impact despite basic tooling. No loss figure is stated.","tags":[],"date_detected":null,"near_miss":null,"root_cause":null,"root_cause_detail":null,"failure_class_detail":null,"responsible_party":null,"adversarial_actor":null,"input_provenance":null,"time_to_contain":null,"operations_paused":null,"remediation":null,"precursor_events":null,"primary_risk":"RK07","secondary_risks":["RK04","RK12"],"control_chain":[{"layer":"L1","party":"provider","party_name":null,"control_id":"D1","outcome":"got_around","reason":null,"quote":"The attacker’s conversations with jailbroken AI tools included getting step by step advice on how to set up a fraudulent Voice over IP (VoIP) operation.","source_index":0,"reviewed":false},{"layer":"L1","party":"provider","party_name":null,"control_id":"D1","outcome":"got_around","reason":null,"quote":"DeepSeek models are inexpensive and easy to access, and readily jailbroken with a simple system-prompt override.","source_index":0,"reviewed":false},{"layer":"L2","party":"harness_vendor","party_name":null,"control_id":"E1","outcome":"got_around","reason":null,"quote":"The Cursor IDE was configured through a proxy script named cipher-proxy.sh that set environment variables to enable full agent mode and bypass its permission and sandbox controls.","source_index":0,"reviewed":false}]},"impact":{"direct_loss_usd":null,"remediation_hours":null,"records_exposed":null,"third_parties_affected_count":null,"severity":"moderate","severity_because":"Moderate test met: data of others (PII on students) touched, with a bounded scope stated (one database of more than 5,200 records).","harm_class":["privacy","third_party_security"],"affected_description":"Targets in India, including universities","protected_class_involved":null,"data_categories_exposed":["personally identifiable information (PII) on students"],"loss_range_usd":null,"reversibility":null,"downstream_exposure_count":null},"evidence":{"sources":[{"url":"https://unit42.paloaltonetworks.com/ai-assisted-cybercrime-analysis/","publisher":"Unit 42","date":"2026-10-21","type":"press","title":"“Next Exploit Opportunities”: How a Threat Actor Used AI to Help Plan and Execute Multiple Types of Cybercrime Activities - Unit 42","archived_url":null,"attachment":"first_report"}],"per_field_confidence":{"system.harness":"C","system.safeguard_state":"C","system.intended_use":"C","impact.harm_class":"C","impact.data_categories_exposed":"C","impact.affected_description":"C","event.incident_class":"D","impact.severity":"D","impact.severity_because":"D","event.primary_risk":"D","event.secondary_risks":"D","event.control_chain":"C","event.summary":"C","system.autonomy_level":"D","system.deployment_context":"D","event.trigger":"D","event.detection_method":"D","context.country":"D","system.developer":"D","event.date_disclosed":"D","context.denominator_available":"D"},"field_citations":{"system.harness":[{"source_index":0,"quote":"An actor operating under the handle BUGSTER configured the Hermes Agent framework with DeepSeek V4 as its reasoning engine and Cursor integrated development environment (IDE) as a secondary tool, both with safety controls disabled."}],"system.safeguard_state":[{"source_index":0,"quote":"An actor operating under the handle BUGSTER configured the Hermes Agent framework with DeepSeek V4 as its reasoning engine and Cursor integrated development environment (IDE) as a secondary tool, both with safety controls disabled."}],"system.intended_use":[{"source_index":0,"quote":"The conversation was with one of several AI assistants that the actor was using to get advice on planning and executing a variety of cybercrime activities."}],"impact.harm_class":[{"source_index":0,"quote":"The actor also achieved a confirmed intrusion against a second university, where they exfiltrated a database of more than 5,200 records, including personally identifiable information (PII) on students."}],"impact.data_categories_exposed":[{"source_index":0,"quote":"The actor also achieved a confirmed intrusion against a second university, where they exfiltrated a database of more than 5,200 records, including personally identifiable information (PII) on students."}],"impact.affected_description":[{"source_index":0,"quote":"Despite some apparent bumbling, the actor had an impact, attacking a number of targets in India, including universities."}],"event.incident_class":[{"source_index":0,"quote":"The actor also achieved a confirmed intrusion against a second university, where they exfiltrated a database of more than 5,200 records, including personally identifiable information (PII) on students."}],"impact.severity":[{"source_index":0,"quote":"The actor also achieved a confirmed intrusion against a second university, where they exfiltrated a database of more than 5,200 records, including personally identifiable information (PII) on students."}],"impact.severity_because":[{"source_index":0,"quote":"The actor also achieved a confirmed intrusion against a second university, where they exfiltrated a database of more than 5,200 records, including personally identifiable information (PII) on students."}],"event.control_chain":[{"source_index":0,"quote":"The attacker’s conversations with jailbroken AI tools included getting step by step advice on how to set up a fraudulent Voice over IP (VoIP) operation."},{"source_index":0,"quote":"DeepSeek models are inexpensive and easy to access, and readily jailbroken with a simple system-prompt override."},{"source_index":0,"quote":"The Cursor IDE was configured through a proxy script named cipher-proxy.sh that set environment variables to enable full agent mode and bypass its permission and sandbox controls."}],"event.summary":[{"source_index":0,"quote":"An actor operating under the handle BUGSTER configured the Hermes Agent framework with DeepSeek V4 as its reasoning engine and Cursor integrated development environment (IDE) as a secondary tool, both with safety controls disabled."},{"source_index":0,"quote":"The actor ran an AI-generated brute-force campaign against an Indian university’s website."},{"source_index":0,"quote":"The actor also achieved a confirmed intrusion against a second university, where they exfiltrated a database of more than 5,200 records, including personally identifiable information (PII) on students."},{"source_index":0,"quote":"The actor also built specialized AI bot profiles to acquire VoIP fraud expertise through conversation and maintained automated pipelines for Wi-Fi password cracking and cryptocurrency whale reconnaissance (i.e., identifying large cryptocurrency holders)."},{"source_index":0,"quote":"Unit 42 identified and investigated this threat actor's operational environment."},{"source_index":0,"quote":"The attacker’s conversations with jailbroken AI tools included getting step by step advice on how to set up a fraudulent Voice over IP (VoIP) operation."},{"source_index":0,"quote":"Despite some apparent bumbling, the actor had an impact, attacking a number of targets in India, including universities."}]},"notes_on_gaps":"Single press source (Unit 42, dated 2026-10-21 in its header); no implicated-party or government report. Not stated: incident date, detection date, duration, time to contain, losses in money, exact record count (only more than 5,200 given, so records_exposed omitted), count of affected third parties, the operator's location, and whether any operations were paused. Developer of DeepSeek V4 not stated in the body, so omitted. Autonomy level, lifecycle stage and deployment context not stated, so omitted. The actor's country is not stated; context.country is set to IN as a judgement from the targets located primarily in India, not the actor's location. Attribution to a person or handle is the source's claim and is not confirmed. Failure class harmful_compliance is a judgement from the jailbroken advisory conversations; data_exfiltration and credential_misuse are secondary judgements from the intrusion and admin credential sentences. Trigger human_misuse is a judgement from the actor directing the tools. No remediation by the affected parties is reported. [case created by the backfill: read by claude-haiku-5-5, backfill-2026-11-01.2, 2026-11-01; not yet reviewed] [2026-11-01: evidence level set: Preliminary (Publisher not yet named · subject to change)]"},"context":{"industry":null,"org_size_band":null,"country":"unknown","denominator_available":false,"denominator":null,"related_incident_ids":[],"systemic_exposure":null,"shared_model_id":null,"upstream_incident_id":null,"incident_evidence":null,"regulatory":null,"coverage":null},"narrative":{"steps":[{"step":"went_wrong","text":"Safety controls were disabled, and jailbroken assistants gave step-by-step advice on setting up a fraudulent VoIP operation.","quote":"The attacker’s conversations with jailbroken AI tools included getting step by step advice on how to set up a fraudulent Voice over IP (VoIP) operation.","source_index":0,"grade":"C"},{"step":"reached","text":"The actor gained a confirmed intrusion into a second university and exfiltrated a database of more than 5,200 records, including PII on students.","quote":"The actor also achieved a confirmed intrusion against a second university, where they exfiltrated a database of more than 5,200 records, including personally identifiable information (PII) on students.","source_index":0,"grade":"C"},{"step":"caught","text":"Unit 42 identified and investigated the actor's operational environment.","quote":"Unit 42 identified and investigated this threat actor's operational environment.","source_index":0,"grade":"C"}],"model":"claude-haiku-5-5","prompt_version":"backfill-2026-11-01.2","extracted_at":"2026-11-01","source_urls":["https://unit42.paloaltonetworks.com/ai-assisted-cybercrime-analysis/"]},"analysis":{"sections":{"timeline":[{"text":"The actor ran an AI-generated brute-force campaign against an Indian university's website.","quote":"The actor ran an AI-generated brute-force campaign against an Indian university’s website.","source_index":0,"when":null},{"text":"The actor achieved a confirmed intrusion against a second university and exfiltrated a database of more than 5,200 records.","quote":"The actor also achieved a confirmed intrusion against a second university, where they exfiltrated a database of more than 5,200 records, including personally identifiable information (PII) on students.","source_index":0,"when":null}]},"model":"claude-haiku-5-5","prompt_version":"backfill-2026-11-01.2","extracted_at":"2026-11-01","source_urls":["https://unit42.paloaltonetworks.com/ai-assisted-cybercrime-analysis/"]}}