30
42
42
35
Sound Debloating of Redundant Checks in Zero-Knowledge Machine-Learning Circuits
Automated framework soundly removes up to 48.7% of redundant constraints in ezkl and zkml ZK-ML circuits, cutting prover time by up to 72.8%.
The framework uses whole-circuit abstract interpretation and a provenance graph to verify that each removed redundant check (range proofs, sign lookups, bit decompositions) remains entailed by the rest of the circuit, provably preserving soundness. It was evaluated on MLP, CNN, RNN, and transformer circuits generated by ezkl and zkml, with up to 25.3 million constraints. It removes up to 48.7% of constraints and reduces prover time by up to 72.8% without weakening security. Under-constrained circuits in deployed ZK systems have previously enabled attackers to forge transactions and bypass verification.
35
Nebulon Enterprise Simulated Threats for Phishing Research (NEST-Phish): A Synthetic Enterprise Phishing Email Dataset for Behavioral and Machine-Learning Research
Researchers release NEST-Phish, a synthetic enterprise phishing email dataset with matched legitimate and phishing emails and cue annotations for detection research.
Academic researchers introduce NEST-Phish, a publicly released synthetic enterprise phishing email dataset built around a fictitious organization named Nebulon. It contains matched synthetic legitimate and phishing emails across a broad set of workplace communication themes, each with interpretable phishing-cue annotations. Human-subject categorizations and supervised classifier evaluations indicate the dataset supports meaningful variation in phishing judgments and provides learnable signal for detection models. It is intended to support work on phishing detection, human susceptibility, explainability, and benchmark development.
35
30
55
42
42
60
42
60
60
55
Hackers Exploit Marimo RCE to Steal AWS Credentials and Reach Bastion Host in 8 Seconds
Attackers exploited pre-auth RCE CVE-2026-39987 in Marimo notebooks to steal AWS credentials and SSH into a bastion host in eight seconds.
Sysdig Threat Research Team documented an intrusion abusing CVE-2026-39987, an unauthenticated RCE in Marimo's terminal WebSocket endpoint affecting versions up to 0.20.4 and fixed in 0.23.0. The attacker harvested AWS credentials from the host environment and Redis backend, queried AWS Secrets Manager to retrieve an SSH private key, and authenticated to an internet-facing SSH bastion host just eight seconds after opening the WebSocket session. Human-operated custom tooling, not an AI agent, executed the full exploit-to-lateral-movement chain.
65
42
30
30
30
60
60
30
30
55
55
30
30
30
57
30
55
57
57
55
57
30
60
30
Attackers Expose Ongoing AI Tool Use Targeting Organizations in Latin America
Unit 42 documents two AI-assisted intrusion campaigns against Latin American government, utility, and financial organizations using LLM-orchestrated tooling.
Palo Alto Networks Unit 42 tracks two ongoing intrusion clusters, CL-CRI-1131 (Mexican transportation, federal ministries, municipal water utilities) and CL-CRI-1163 (Brazilian financial sector), both using living-off-the-land techniques, SOCKS5 relays, and custom RATs. The attackers appear to orchestrate operations via commercial LLMs like Claude and GPT-4.1, evidenced by iterative batch scripts and AI-generated tunneling tool naming. The Mexican campaign (also reported as Operation Escaneo by CloudSEK) exfiltrated sensitive data via dynamic-DNS infrastructure with rotated multi-SAN TLS certificates between February and June 2026. This signals broader adoption of AI-enhanced operations by diverse threat groups in the region.
72