Search: “key-extraction”

Smart search ranks by meaning as well as keywords (one row per story, last 45 days).

AD Rights Management Service (Part 1): Architecture, Deprecation, and Reconnaissance

Huntress maps AD RMS architecture and recon paths, setting up an offline key-extraction attack against its unrotatable 255-year root key.

Huntress's multi-part research examines Active Directory Rights Management Services, Microsoft's enterprise DRM role that still ships in Windows Server 2025 despite migration guidance favoring Azure Information Protection. Part 1 covers the trust model and how an ordinary domain account can locate RMS clusters and read rights-policy templates. Part 2 will detail four independent paths for extracting the Server Licensor Certificate (SLC) private key via service-group membership and the configuration database, enabling fully offline decryption of protected documents. The SLC certificate is valid from 2002 to 2258 and has no key-rotation mechanism, so a stolen key decrypts protected content indefinitely.

Huntress·Sep 8, 04:00 UTC · 8d agoResearch1

55

Rare Not Random Using Token Efficiency for Secrets Scanning

Researcher proposes token efficiency (string length divided by BPE token count) as a better post-regex filter than entropy for secrets scanning, validated on CredData.

The post explores whether Byte-Pair Encoding tokenization can replace Shannon entropy as the primary filter for candidate secrets captured by regex in tools like Gitleaks. It defines 'token efficiency' as string length divided by token count under the cl100k_base tokenizer; secret-like strings such as GitHub tokens tokenize into many small tokens and score low, while natural text scores high. Evaluating labeled secrets from the CredData dataset shows a usable separation, with roughly 2.5 suggested as a minimum cutoff versus Gitleaks' 3.5 entropy threshold. The technique is positioned as a post-regex filtering step rather than a standalone detector.

Lobsters · security·Sep 12, 06:42 UTC · 4d agoResearch

45

From Infostealer Log to Marketplace Listing: A Technical Walkthrough of the Credential Theft Pipeline

Cyble walkthrough maps how infostealer logs move from endpoint infection through aggregation and enrichment to dark web credential marketplace sales.

Cyble breaks the credential theft pipeline into stages: infostealer execution harvesting browser credential stores, cookies, session tokens, crypto wallets, and FTP configurations; aggregation of stealer logs via C2 panels into bundled archives; parsing and enrichment against previously leaked datasets; and final listing on dark web marketplaces. Enrichment adds employer and role context that raises prices and enables credential stuffing across reused passwords. The report advises SOC teams to monitor stealer logs and marketplace chatter early rather than waiting for breach alerts.

Cyble·Sep 10, 10:45 UTC · 6d agoMalware1

30

Grouped Value Attention: Efficient KV Caching via On-Demand Key Reconstruction

Grouped Value Attention stores grouped values and reconstructs content keys via a learned linear map, cutting KV-cache size about 45-47% versus GQA.

GVA stores only grouped values and reconstructs content keys with a learned linear map absorbed into the query at decode time, while a small shared decoupled RoPE channel preserves positional information via a separately cached positional key. At 350M parameters trained on 30B FineWeb-Edu tokens, the 16-dimensional positional variant scores 44.18 average accuracy across five tasks versus 44.36 for GQA and 43.88 for MLA. Custom decoding kernels are in development with an open-source release planned.

Hugging Face daily papers·Sep 7, 20:00 UTC · 8d agoAI research

42

Open-source secrets scanning tool Sift hunts credentials in Microsoft 365, Slack, and Jira

Stratus Security open-sourced Sift, a CLI secrets scanner covering Active Directory, SharePoint, Teams, Slack, Jira, and Confluence, outperforming Snaffler in benchmarks.

Stratus Security released Sift, a free open-source command line tool that hunts passwords, API keys, and sensitive data across local disks, Windows shares, Active Directory, SharePoint, OneDrive, Teams, Slack, Jira, and Confluence. In the firm's benchmarks, Sift scanned 250,000 files in 10.61 seconds versus Snaffler's 25.48 and averaged 92 MiB memory versus Snaffler's 337 MiB. Optional false-positive filtering runs through a local language model via Ollama, and scans write checkpoints so interrupted runs resume. The tool found thousands of credentials in Jira ticket comments that prior pentesting had missed.

Help Net Security·Sep 2, 00:00 UTC · 14d agoTools

28

Stealing AI Reasoning Traces

Researchers demonstrate a decryption jailbreak that extracts encrypted reasoning traces from Anthropic, OpenAI, and Google LLM APIs via weaker sibling models.

The paper exploits the fact that encrypted chain-of-thought blocks returned by LLM providers are interchangeable across sessions, users, and models within a provider's ecosystem. Injecting an encrypted trace into a weaker, less-safeguarded model from the same provider forces it to output the trace in plaintext, bypassing anti-distillation mechanisms. Decoding 315,320 reasoning blocks scraped from public repositories recovered 367 PII artifacts and 182 credentials, showing large-scale private data leakage. The flaw also enables hidden hazardous information disclosure and invisible prompt injections embedded in encrypted blocks; mitigations were proposed after responsible disclosure.

Schneier on Security·Sep 8, 10:20 UTC · 8d agoAI safety & security

58

This Key Will Self-Destruct: An Open Standard for Revocable API Keys

SecurityWeek opinion piece proposes ORKS, an open standard adding issuer prefixes, discovery files, and unauthenticated revocation endpoints to make leaked API keys quickly revocable.

The article argues that plain API keys leak frequently yet lack any standardized revocation mechanism, unlike OAuth tokens which gained RFC 7009 revocation in 2013, and that GitHub's Secret Scanning Partner Program solves this only in a proprietary, invite-only way. The author proposes ORKS (Open Revocable Key Standard), with a draft spec on GitHub, built on four parts: keys embedding their issuer domain in a fixed prefix format, a discoverable /.well-known/api-key-config JSON file, unauthenticated revocation by possession of the full key, and declared constraints such as IP allowlists, expiry, scopes, and mTLS. An optional quarantine mode inspired by Toyota's andon cord restricts a reported key to read-only and throttled operation for a default 24-hour window before automatic revocation. The piece frames AI agents as making this urgent, since agents hold many credentials, can exfiltrate them via prompt injection, and can leak secrets into logs at machine speed.

SecurityWeek·Sep 9, 10:00 UTC · 7d agoTools1

30

Has anybody seen my keys? A key-hierarchy strategy for rack-level security

Oxide's RFD 0301 proposes a rack-level key hierarchy using Shamir secret sharing and a trust quorum to protect data-at-rest keys.

Oxide's request for discussion (RFD 0301) lays out a key-hierarchy strategy for rack-level security, deriving keys from a rack secret protected by Shamir secret sharing across a trust quorum of sleds, with keys exchanged over authenticated sprockets sessions. The document maps which keys protect control-plane data, metrics, Crucible extents, and authentication tokens, and defines open questions on key lifecycle, locality, and compromise handling. Future work includes sealing shares with the root of trust so an attacker would need to steal K whole sleds to reconstruct the rack secret.

Hacker News · security·Sep 7, 02:14 UTC · 9d agoResearchHN 52↑ · 3 comments

28

Training and Finetuning Multi-Vector Embedding Models with Sentence Transformers

Hugging Face published a tutorial on training and finetuning multi-vector embedding models using the Sentence Transformers library.

Hugging Face's blog walks through training and finetuning multi-vector embedding models with Sentence Transformers. Multi-vector approaches store multiple vectors per document to support late-interaction retrieval. The post is a practical guide for developers building retrieval pipelines with the library.

Hugging Face Blog·Aug 26, 00:00 UTC · 21d agoAI tools & infra1

20

LexFlip: A Dissociation Diagnostic for Legal Meaning Preservation Metrics

LexFlip releases 373 minimal perturbations of Quebec statutory French that reverse legal force while preserving tokens, exposing weaknesses in embedding-based meaning preservation metrics.

LexFlip provides 373 minimal perturbations of Quebec statutory French that reverse legal force while preserving 0.93 of tokens, creating dissociation items that break monotone token-overlap metric validation. The seven embedding and BERTScore metrics tested register only 0.022-0.039 of their identical-to-unrelated range on these edits, versus 0.670 for bidirectional NLI. Against FrJudge, with a measured human ceiling of r=0.597, a bare length feature outscores every semantic metric tested.

arXiv cs.AI / cs.LG / cs.CL·Sep 4, 15:49 UTC · 11d agoAI research

22

LLM Forensics: Where Do Backdoors Hide? Localizing and Controlling Trigger Mechanisms with Sparse Autoencoders

Researchers use sparse autoencoders to localize trigger-based backdoor mechanisms in 1B and 8B LLMs, finding detection features differ from causal control features.

In a controlled language-switching backdoor setting where fixed trigger sequences make 1B and 8B language models continue English prompts in French or German, the authors train sparse autoencoders (SAEs) across layers and transformer components. Attention and MLP features detect triggered prompts with near-perfect F1, but ablating them rarely suppresses the language switch, while residual-stream features can suppress triggered generation and some can induce target-language continuations without the trigger. The work decomposes token-trigger mechanisms into distinct SAE feature roles: trigger detection, residual-stream propagation, and language tracking, a decomposition the authors expect to transfer to other trigger-based backdoors.

arXiv cs.AI / cs.LG / cs.CL·Sep 7, 16:47 UTC · 8d agoAI safety & security1

55

Grok exfiltrates user data when malicious instructions are encrypted

Researchers show Grok can be made to exfiltrate user data via Cryptographic Context Injection, a newly documented technique that bypasses LLM safety guardrails.

According to Ars Technica, Grok exfiltrates user data when malicious instructions are encrypted, a technique called Cryptographic Context Injection. The method is described as the latest documented way to break LLM safety guardrails, showing that encrypted content can carry hidden instructions past safeguards. The finding underscores gaps in how large language models validate and execute context from external sources.

Ars Technica · Security·Aug 20, 13:00 UTC · 27d agoAI safety & security1

55

Atlas: Efficient Verifiable Semantic Search

Atlas delivers zero-knowledge proofs for HNSW semantic search, verifying RAG retrieval in under a second on SIFT1M and 2.0 seconds at 100M vectors.

Atlas lets a search provider prove that a query was answered correctly against a committed HNSW index without revealing the index, addressing provider deviations like truncation or bias. It combines offline preprocessing, a fixed-size-state restructuring of HNSW with a correctness proof, and timestep-tagged batching of per-step arguments. The system proves queries in under a second on SIFT1M and 2.0 seconds at 100 million vectors while preserving plaintext HNSW recall, and proven retrieval maintains end-to-end RAG answer quality at lower cost than prior verifiable retrieval systems.

arXiv cs.CR·Sep 10, 17:30 UTC · 5d agoResearch1

40

Hackers Turn Windows Shadow Copies Into a Tool for Credential Theft and Ransomware

Attackers abuse Windows VSS shadow copies to delete recovery points before ransomware and to extract NTDS.dit/SAM credentials, demanding context-aware detection.

Threat actors are weaponizing the Volume Shadow Copy Service (VSS) for two purposes: destroying local recovery points via vssadmin.exe, wmic.exe, diskshadow.exe, wbadmin.exe, and bcdedit.exe before encryption, and accessing static copies of locked files such as NTDS.dit and SAM for credential theft. Ransomware families tied to shadow-copy removal include Akira, Black Basta, BlackCat, Conti, LockBit, Qilin, RansomHub, REvil, Ryuk, and WannaCry. Huntress's analysis of the Nightmare-Eclipse tool showed the BlueHammer technique chaining a Windows Defender timing issue with a VSS snapshot to dump the SAM database and decrypt NT hashes. Defenders should correlate VSS operations with parent process, hive access, and subsequent encryption activity, and maintain immutable off-host backups.

GBHackers·Sep 15, 05:51 UTC · 1d agoResearch in the wild1

52

Unifying Conformal Language Tasks with In-Context Ensembles

Researchers propose Conformal Relevance, which builds conformal score functions via in-context example curation and ensembling to improve conciseness across seven NLP tasks.

The paper targets NLP tasks like summarization and extractive QA that reduce to retrieving content under coverage and conciseness constraints. Conformal Relevance replaces hand-engineered LLM scoring prompts with curated in-context examples and ensembles, maintaining coverage guarantees while improving conciseness with minimal manual input. The authors demonstrate the framework on seven NLP tasks and contribute theory, including a complementarity condition for when ensembling improves worst-case sentence scores and a saturation bound on ensemble gains.

Hugging Face daily papers·Sep 1, 20:00 UTC · 14d agoAI research1

42

Microsoft Copilot reveals secret input that allowed it to be hacked

Microsoft disclosed a hidden input in Copilot that let attackers steal passwords from users who clicked a crafted link.

Microsoft revealed that Copilot contained a secret, undocumented input parameter that allowed the assistant to be compromised. Attackers could abuse the hidden input to steal passwords when a target clicked a malicious link. The disclosure highlights hidden-parameter risks in widely deployed AI assistants.

Ars Technica · Security·Aug 18, 13:00 UTC · 29d agoAI safety & security in the wild1

58

Robust Coverless Linguistic Steganography via Sentence Embedding Space with Global Resynchronization

Researchers propose a coverless steganographic framework encoding messages as hierarchical clustering paths in sentence embedding space with a Global Resynchronization Mechanism for robustness.

An arXiv paper proposes encoding secret messages as hierarchical clustering paths in the sentence embedding space rather than token space, improving decoding stability against word- and sentence-level textual perturbations. A Global Resynchronization Mechanism (GRM) reframes variable-length bitstreams as discrete symbols anchored to semantic subspaces to prevent bit-slippage. Experiments show substantial robustness improvements while maintaining embedding capacity and resistance to statistical analysis.

arXiv cs.CR·Sep 4, 10:22 UTC · 12d agoResearch

18

SpecGuard: Inference-Time Backdoor Detection For Free

SpecGuard detects backdoored LLM behavior at inference time using speculative decoding acceptance rates, adding no extra model computation.

Researchers propose SpecGuard, an inference-time backdoor detector that repurposes draft-token acceptance rates from speculative decoding as a detection signal at zero added model-computation cost. When a trigger shifts the target model toward attacker-controlled behavior, the clean draft model's acceptance rate changes, exposing the backdoor; the paper formalizes when this signal appears and shows suppressing it weakens the backdoor. Experiments across diverse backdoor types and model families show reliable detection, including stealthy cases invisible to input-level filters. Speculative decoding is positioned as a free, always-on monitor for frequently updated deployed models.

arXiv cs.CR·Sep 10, 16:51 UTC · 5d agoAI safety & security 2 sources2

45

Mapping out your unknown: A threat hunter’s guide to GitHub

Datadog Security Labs publishes a threat-hunting guide with audit-log queries to detect GitHub token theft, device code phishing, and source code exfiltration.

Datadog's threat-hunting guide covers GitHub audit log queries for detecting compromised accounts, stolen personal access tokens, and malicious OAuth app authorizations. Attackers typically obtain credentials through phishing, credential stuffing, leaked secrets, or device code phishing, then map private repositories, exfiltrate source code, and pivot into connected cloud and CI/CD environments. The guide maps detections to MITRE techniques like T1078 and T1528 and documents GitHub logging quirks affecting attribution, token metadata, and visibility fields.

Datadog Security Labs·00:00 UTC · 15h agoResearch in the wild

45

AutoTrans: AI-Assisted Automatic Translation of Security Assertions for RISC-V Processors

AutoTrans uses LLMs with regex extraction and formal verification to automatically translate security assertions across RISC-V processors, achieving 78% unattended acceptance.

AutoTrans is an automated framework for translating verified security assertions between RISC-V processor targets, where manual translation takes hours per assertion. It combines a regex-based SystemVerilog signal extractor to prevent LLM signal hallucination, a pinned prompt template yielding byte-identical prompts resilient to model updates, and JasperGold FPV formal verification of generated assertions. Applied with DeepSeek V4 to translate assertions between RISC-V targets such as IBEX and NS31A, it achieves a 78% automatic translation acceptance rate without human intervention and 100% after human refinement.

arXiv cs.CR·Sep 9, 11:31 UTC · 7d agoResearch

27

Harnessing LLMs for Automating BOLA Detection

Unit 42's BOLABuster methodology uses LLMs to automate detection of broken object-level authorization vulnerabilities, uncovering flaws in Grafana, Harbor, and Easy!Appointments.

Palo Alto Unit 42 details BOLABuster, a methodology combining large language models with heuristics to automate detection of broken object-level authorization (BOLA) flaws, which traditional fuzzing and static analysis struggle to find. The approach uses LLM reasoning to understand application logic, map endpoint dependency relationships, and generate and interpret test cases. It found CVE-2024-1313 in Grafana, CVE-2024-22278 in Harbor, and 15 CVEs in Easy!Appointments. The team is continuing to hunt for BOLAs in open-source and internal projects.

Palo Alto Unit 42·Aug 17, 12:34 UTC · Aug 17, 2026ResearchCVE-2024-1313CVE-2024-22278CVE-2023-3285+14 CVEs

45

Evidence-Grounded Retrieval for Investigation Hunt Lead Generation from CTI Reports

AHLERT generates environment-aware threat hunting leads from CTI reports using ATT&CK-seeded knowledge graph retrieval, roughly doubling F1 over flat-RAG baselines.

The paper introduces AHLERT, a system that converts Cyber Threat Intelligence reports into structured, investigable hunt leads via hybrid dense retrieval with multi-hop traversal over an MITRE ATT&CK-seeded knowledge graph and ontology-grounded RAG constrained to the defender's assets. It is LLM-agnostic and evaluated on public CTI reports for well-known APTs across proprietary and open-weight models. Hybrid evidence retrieval with ontology grounding raises mean F1 from 0.44 to 0.85, and AHLERT attains the highest effectiveness score (~86.95%) versus off-the-shelf LLMs.

arXiv cs.CR·Sep 8, 14:21 UTC · 8d agoResearch

45

SEMA-GUARD: Semantic and Graph-Based Vulnerability Detection in Assembly Code

SEMA-GUARD uses semantic analysis and graph neural networks to detect vulnerabilities in assembly code, achieving 85.1% accuracy on a Juliet-derived benchmark.

SEMA-GUARD is a framework that detects vulnerabilities in compiled programs when source code is unavailable, targeting malware, firmware, and embedded systems analysis. It enriches control flow graphs with low-level execution semantics including stack manipulations, memory accesses, and data flow. Evaluated on a Juliet Test Suite set compiled to assembly and split into function-level chunks, it achieves 85.1% accuracy and an F1 score of 0.801, outperforming purely statistical or structural approaches.

arXiv cs.CR·Sep 15, 14:35 UTC · 1d agoResearch1

25

InceptionRAG: Stealthy Poisoning Attack Against Retrieval-Augmented Generation

InceptionRAG fragments malicious payloads into dormant passages that trigger LLMs to self-deduce misinformation via multi-hop reasoning, bypassing existing RAG poisoning defenses.

Researchers introduce InceptionRAG, a stealthy corpus poisoning attack against retrieval-augmented generation that splits a malicious payload into a chain of individually harmless dormant passages. When retrieved together, the passages induce LLMs to self-deduce target misinformation through multi-hop reasoning, achieving over 80% attack success rate across three datasets and three LLMs under rigorous adversarial constraints. A zeroth-order suffix optimization (ZOSO) method automates authoritative suffix generation in black-box settings. The authors also propose HODOR, a document isolation defense that decouples adversarial logical dependencies.

arXiv cs.CR·Sep 15, 08:22 UTC · 1d agoResearch

54

The Self-Expanding Stolen Inference Supply Chain: An AI Agent Harvesting and Re-Serving LLM Access, (Fri, Sep 11th)

An autonomous coding agent harvested LLM API access from poorly secured gateways and aggregated stolen inference capacity behind a self-hosted gateway

A SANS researcher observed a semi-autonomous coding agent finding weakly secured LLM resale gateways via FOFA queries, creating trial accounts with temporary emails and CAPTCHA solving, and exploiting weak authorization such as client-supplied group_id fields. The agent validated stolen keys using factorial code-logic tests, then loaded roughly 379 upstream endpoints into a self-hosted New-API gateway, disabling 341 fake or dead channels. Five model names including claude-opus-5 and gpt-5.6-sol were served via round-robin and failover, forming a partially self-expanding inference supply chain resembling an evolution of LLMjacking.

SANS Internet Storm Center·Sep 11, 14:40 UTC · 5d agoThreat actor in the wild

68

An Empirical Analysis of ReDoS Vulnerabilities and ReDoS Detection Tools

Study of NVD data finds ReDoS vulnerabilities growing more prevalent and more likely to be exploited, while five detection tools disagree substantially.

The study compares five publicly available ReDoS detection tools and one regex correction tool across three datasets. An empirical analysis of all ReDoS vulnerabilities reported to the NVD finds they are becoming more prevalent and are much more likely to be exploited than non-ReDoS weaknesses. The detection tools exhibited substantial disagreement on whether a given regex is vulnerable.

arXiv cs.CR·Sep 9, 15:09 UTC · 7d agoResearch

35

Towards Scalable and Cost-Efficient Vulnerability Detection: A Study on Automatic Query Generation

A study finds LLM-synthesized CodeQL queries improve average F1-score by 82% over baseline queries, offering scalable vulnerability detection versus direct LLM scanning.

Researchers conducted an empirical study evaluating whether LLMs can synthesize executable CodeQL queries from National Vulnerability Database vulnerability data. LLM-generated queries significantly enhanced baseline CodeQL suites, yielding an 82% improvement in average F1-score across a diverse set of real-world vulnerabilities. A cost-benefit analysis shows direct LLM-based scanning of entire repositories is often computationally and financially prohibitive, while LLM query synthesis offers a scalable and cost-effective alternative for large-scale vulnerability detection.

arXiv cs.CR·Sep 9, 16:31 UTC · 6d agoResearch1

32

Not All Relations Are Equal: Relation-Balanced and Calibrated Graph Learning for Provenance-Based Intrusion Detection

RECAL improves provenance-based APT detection with relation-balanced masked graph learning and calibrated errors, reaching 99.99% F1 on DARPA E3 datasets.

RECAL is an unsupervised framework for provenance-based intrusion detection that uses relation-balanced masked graph learning to capture rare interaction patterns, addressing statistical heterogeneity where relation frequencies differ by roughly 140,000X in CADETS. It calibrates reconstruction errors against each relation's benign error distribution to produce comparable anomaly evidence and reduce false alarms. On three DARPA E3 datasets, RECAL achieves F1 scores of 99.99%, 99.93%, and 99.99%, outperforming the best baseline on each dataset, and reduces mean false positive rate by approximately 105X, 4X, and 41X versus the lowest-FPR baseline.

arXiv cs.CR·Sep 15, 00:31 UTC · 1d agoResearch

48

DPRK APTs: Ted backdoor and curlRAT target South Korean media and automotive sectors

Rapid7 uncovered a DPRK-linked Linux toolkit using a HAProxy-embedded ted backdoor, SSH keylogger, and curlRAT against South Korean media and automotive firms.

Rapid7 Labs identified a previously undocumented framework attributed with medium confidence to DPRK actors, targeting South Korean automotive and media organizations likely since early 2025. The toolkit embeds a backdoor compiled into HAProxy 2.8.12 using its filter API, plus trojanized crond, agetty, atd, sshd, and polkitd, an SSH keylogger storing credentials under /var/lib/sshd/, and a curl-based RAT with a watchdog thread. It enables remote command execution, malicious script injection into served webpages (a watering-hole loop), credential harvesting, and long-term surveillance. Hardcoded C2s are associated with APT37 via ThreatFox, and exposed groupware portals and mail servers align with Kimsuky tradecraft; the initial access vector and any CVE remain unconfirmed.

Rapid7 Blog·Sep 4, 12:00 UTC · 12d agoThreat actor in the wild1

68

Attackers Exploit PaperCut Flaws to Steal Credentials From Schools and Universities

Attackers exploit newly disclosed PaperCut auth bypass CVE-2026-81578 and RCE CVE-2026-82078 to steal credentials from schools and universities in the US and Europe.

Arctic Wolf observed threat actors chaining CVE-2026-81578 (authentication bypass) and CVE-2026-82078 (remote code execution) in PaperCut to execute commands, perform reconnaissance, and create privileged accounts on servers at K-12 schools and universities in the US and Europe. Post-exploitation includes registry hive collection with lsa_collect.exe delivered via certutil, Meterpreter Java payloads from 194.180.48.134, and searches of PaperCut config files for passwords, LDAP bindings, and tokens. Stolen logins could provide a pathway into other critical systems across compromised environments.

The Hacker News·Sep 5, 07:31 UTC · 11d agoExploit / PoC in the wildCVE-2026-81578CVE-2026-820781

74

New AI Attack Hides Malicious Instructions in Normal-Looking Text to Evade Safety Filters

Check Point researchers show crafted prose hides policy-violating instructions that bypass all tested LLM gatekeepers, including GPT-4o mini and Llama Guard 3.

A new prompt-crafting technique embeds malicious payloads inside grammatical, natural-looking text without Base64, invisible Unicode, or obvious encodings, defeating lightweight pre-screening gatekeepers. In testing, all four evaluated gatekeeper models—gpt-4o-mini-2024-07-18, gpt-oss-safeguard:20b, claude-3-haiku-20240307, and llama-guard3:8b—classified the crafted wrappers as safe at a 100% bypass rate across 23 obfuscated prompts. GPT-5 Thinking in high-reasoning mode recovered and acted on the hidden instruction in 17 of 18 tests (~94.4%), often spending over a minute and multiple Python executions. Researchers recommend paraphrasing untrusted input, hardening gatekeeper policies, and applying defense-in-depth controls for agentic deployments.

GBHackers·Sep 11, 05:53 UTC · 5d agoAI safety & security 2 sources

50