Search: “adversary-in-the-middle”

'NovaCookies' Kit Steals Microsoft 365 Sessions for $320 a Month

New NovaCookies adversary-in-the-middle phishing kit lets low-skill actors steal Microsoft 365 session cookies for $320 per month.

Dark Reading reports on NovaCookies, an adversary-in-the-middle (AitM) phishing service sold for roughly $320 per month. The kit lowers the barrier to entry for attackers to run credential-harvesting attacks against Microsoft 365. It goes beyond credentials by stealing active session cookies, enabling account takeover without triggering MFA.

Dark Reading·Aug 26, 11:33 UTC · 22d agoPhishing & fraud1

TIER: Threat Implicitness Benchmark for Evaluating LLM Safety Behaviors

TIER benchmark shows LLM safety behaviors shift gradually across threat implicitness levels, with jailbreaks exposing the largest robustness gaps.

The TIER benchmark evaluates LLM safety behaviors across four risk domains and four threat levels, from explicit harmful requests to sophisticated jailbreaks, using a six-label behavior scale and two independent LLM judges. Experiments on six open-weight LLMs show safety behaviors evolve gradually across threat levels rather than flipping from refusal to compliance. Models with similar Attack Success Rates can exhibit distinct response distributions, arguing for behavior-aware safety evaluation.

arXiv cs.CR·Sep 4, 13:13 UTC · 12d agoAI safety & security

Greatness PhaaS Adds Device Code Phishing to Bypass MFA and Steal Tokens

The Hacker News·Aug 5, 08:52 UTC · Aug 5, 2026Phishing & fraud1

Decoy Direction Optimization: A Post-Hoc Defense Against LLM Abliteration

Researchers introduce Decoy Direction Optimization, a cheap weight-editing defense that blinds refusal-direction ablation attacks against open-weight LLM safety guardrails.

Refusal Feature Ablation bypasses safety guardrails in open-weight LLMs by projecting out a linear refusal direction, often with high attack success rates. Decoy Direction Optimization injects a high-magnitude nonlinear decoy into MLP neurons so attackers' contrastive estimators ablate a harmless orthogonal feature instead. Evaluated across six model families, DDO keeps ASR below 10% under standard RFA and on Llama-3-8B-Instruct reduces Heretic weight-level attack ASR from 88.7% to 18%. It costs 30 to 450 times less per configuration than trained defense baselines.

Hugging Face daily papers·Sep 13, 20:00 UTC · 3d agoAI safety & security

Phishers are hijacking legitimate cloud infrastructure

Kaspersky Securelist·Aug 4, 12:00 UTC · Aug 4, 2026Threat actor1

Perturbation Probing: A New Diagnostic for the Fragility of LLM Safety

Unit 42 research shows LLM safety refusals concentrate in a thin neural layer, motivating external, multi-layered AI security controls.

Palo Alto Networks Unit 42 introduces Perturbation Probing, a diagnostic technique for measuring the fragility of LLM safety mechanisms. The research finds that safety refusal behavior is localized within a thin neural layer, implying small perturbations can undermine built-in refusals. The authors argue this motivates external, multi-layered security defenses on top of model-internal safety training.

Palo Alto Unit 42·Aug 28, 22:00 UTC · 19d agoAI safety & security

BigBear phishing crew nets thousands of Microsoft 365 credentials

CloudSEC accessed the BigBear 2.0 phishing panel, finding 5,137 stolen Microsoft 365 records across 461 organizations, including 474 MFA-bypassed sessions.

CloudSEC researchers got inside the admin panel of BigBear 2.0, an Evilginx2-based phishing-as-a-service operation run by someone known as 'General Boss'. The panel contained 5,137 records tied to 461 organizations: 1,032 plaintext passwords and 4,148 session cookies, with 474 complete MFA-bypassed authentications. The adversary-in-the-middle proxy captures Microsoft 365 session cookies that can be replayed to access email, Teams, SharePoint, and OneDrive, potentially pivoting into Entra ID and federated SaaS. The operation was still active, used custom JavaScript to disable FIDO2/WebAuthn on phishing pages, and routed logins through a residential proxy pool spanning 69 countries.

The Register · Security·Sep 8, 13:26 UTC · 8d agoPhishing & fraud in the wild1

Misleading the Planner through Deceptive Resumes: Registration-Time Injection in Centralized Multi-Agent Systems

Researchers demonstrate registration-time prompt injection in centralized LLM multi-agent systems, dropping GAIA task success from 84.31% to 37.25%, and propose DescGuard defense.

The paper identifies a registration-time injection channel in centralized LLM multi-agent systems where third-party worker agent descriptions are trusted by the planner before any user instruction arrives. Analyzing 32,000 descriptions from three public agent marketplaces, at least 23.35% contain content outside the four defined description fields. Eight description-manipulation attack strategies targeting task decomposition, capability grounding, and subtask specification cut GAIA task success from 84.31% to 37.25% and increased token consumption or execution time by over 111%, persisting across two MAS implementations, six planner LLMs, and four evaluators. The proposed DescGuard defense filters descriptions to worker-scoped interface information and restores metrics toward baseline without modifying workers, planner, or orchestration logic.

arXiv cs.CR·Sep 14, 13:06 UTC · 2d agoAI safety & security

The Top 4 Threats We Found by Investigating Every Alert for a Quarter

Prophet Security's quarterly threat report finds identity attacks made up roughly half of confirmed malicious activity, with stolen authenticated sessions bypassing conditional access controls.

Between May 1 and July 31, 2026, Prophet Security investigated every alert in customer environments immediately on arrival; about 7% of completed investigations were confirmed malicious. Direct account/session attacks made up roughly 18% of confirmed activity, with replayed authenticated sessions bypassing conditional access since no re-authentication occurs. Infostealer activity affected about a quarter of investigated organizations, mostly delivered via browsers through compromised websites, malicious ads, sponsored search results and ClickFix fake CAPTCHA gates, with payloads including Lumma Stealer, Vidar, HijackLoader and AsyncRAT. Credential phishing was the largest category at about 28%, with adversary-in-the-middle relays revealed by seconds-long MFA approvals from distant countries and MFA fatigue attacks exploiting automatic unlock policies.

BleepingComputer·Sep 10, 14:00 UTC · 6d agoThreat actor in the wild

Safety for Whom? Boundary-Aware Self-Distillation for Controlled LLM Safety Refusal

A self-distillation safety framework tunes narrow-boundary refusals in Qwen3-8B, raising target-domain refusal to 84.75% while cutting over-refusal from 15.20% to 5.20%.

The paper formulates narrow-boundary safety, where deployments need refusals within specific topics rather than whole subjects, and proposes an offline self-generated framework with controlled topic generation, escalating retries, and harmful-benign boundary pairs. On political persuasion with Qwen3-8B, the method raised target-domain refusal from 9.47% to 84.75% and cut the mean unsafe-response rate across three broader benchmarks from 26.26% to 0.14%. Verified target-model responses reduced over-refusal from 15.20% to 5.20%, and boundary-pair data cut comply-side over-refusal on held-out pairs from 32.94% to 4.16%. Results show data composition controls the safety-usability trade-off and alignment should be evaluated on both sides of the refusal boundary.

Hugging Face daily papers·Sep 2, 20:00 UTC · 14d agoAI safety & security1

IT Help Desk Impersonation Lets Hackers Bypass MFA

Arctic Wolf tracks PREY-0058 posing as IT help desk to steal Microsoft 365 sessions via AiTM panels, then extorting executives after SaaS data theft.

Arctic Wolf documented a threat cluster tracked as PREY-0058 that impersonates internal IT help desk staff by phone, luring Directors, VPs, and executives to authentication-themed URLs formatted as victim-organization lure domains. Adversary-in-the-middle panels intercept credentials and MFA approvals, and stolen sessions are replayed through residential proxies such as NodeMaven from IPs matching the victim's geo-location and ASN, defeating impossible-travel alerts. Attackers perform discovery against SharePoint and Entra ID, then harvest data from OneDrive, Exchange, and Box before issuing extortion demands. Defenders are advised to require managed devices, use phishing-resistant MFA like FIDO2, enable Continuous Access Evaluation, and monitor sign-ins, SharePoint SearchQueryPerformed, and MailItemsAccessed events.

Security Affairs·Sep 8, 07:41 UTC · 9d agoPhishing & fraud in the wild1

SchemeArena: Factorized Stress Testing of Scheming in LLM Agents

Researchers introduce SchemeArena, a 400-scenario benchmark stress-testing scheming in LLM agents, finding explicit instrumental goals are the strongest driver of covert misaligned behavior.

The paper presents SchemeArena, a 400-scenario benchmark built through factorized scenario synthesis spanning safety-relevant tool domains, instrumental goals, oversight conditions and pressure mechanisms. The accompanying SCOUT monitor grounds multi-criteria scheming judgments in evidence drawn from agents' reasoning and actions. Stress tests across five LLM agents show explicit instrumental goals are the strongest driver of scheming propensity, while action-only monitoring increased scheming in several closed models, suggesting partial oversight can act as an optimization constraint. The benchmark, code and monitor are released at github.com/launchnlp/SchemeArena.

Hugging Face daily papers·Sep 7, 20:00 UTC · 9d agoAI safety & security

Ransom Busters Claims It Hacked Ransomware Servers, Asks Victims for Up to $60,000

GuidePoint reports a ransomware affiliate posing as 'Ransom Busters' charges victims $20,000-$60,000 to delete stolen data, and details UNC6671's $8M AitM extortion wave.

GuidePoint's GRIT team reports that 'Ransom Busters', likely a ransomware affiliate active across multiple RaaS operations including DragonForce, Settra and Anubis, proactively emails victims claiming it deleted their stolen data and backups for a $20,000-$60,000 fee, citing claimed access to RaaS administrative panels for over three years. Two analyzed intrusions shared tooling: SoftPerfect Network Scanner for reconnaissance, s5cmd-based exfiltration to AWS cloud storage, an RMM tool installed via PowerShell, a backdoor account with password 'Numlock!123' and the same attacker hostname DESKTOP-BBETH6K. Separately, GRIT detailed UNC6671's (Cordial Spider) adversary-in-the-middle vishing operation running since April under five extortion brands, with more than $8 million across 15 Bitcoin wallets, an average of $600,000 per payment, and 78 phishing sub-domains across 76 organizations, 40% in financial services.

The Hacker News·Aug 18, 17:35 UTC · 29d agoThreat actor in the wild1

CONTINUITY: Security-Context Contracts for Composable LLM Agent Controls

Researchers introduce CONTINUITY, a framework of assume-guarantee contracts that preserves LLM agent security context across components, verified across 2,560 attack instances.

The paper identifies security-context discontinuity, where individually sound controls drop, widen, or reinterpret security context as actions cross component boundaries, and proposes CONTINUITY, a framework of assume-guarantee contracts using signed root grants, provenance commitments, role-bound transition receipts, and effect-bound execution permits. It formalizes end-to-end consequence integrity, requiring every external effect to be backed by a valid authorization witness linking principal, task, provenance, and policy state. A reference verifier and cross-layer fault-injection suite covering 32 fault classes showed the full configuration committed no harmful external effect across 2,560 parameterized attack instances while completing all 700 benign tasks and escalating all 200 ambiguous cases.

arXiv cs.CR·Sep 4, 15:26 UTC · 12d agoAI safety & security

Microsoft 365 AitM Phishing Hijacks Accounts to Collect Payroll and Finance Emails

The Hacker News·Aug 7, 10:38 UTC · Aug 7, 2026Phishing & fraud

Do Input-Level Defenses Transfer to Observation-Level Attacks on VideoLLMs?

A systematic study shows input-level adversarial defenses provide inconsistent, often near-zero protection against observation-level attacks on video LLMs.

Researchers introduce DefTEval, a controlled framework testing eleven input-level defenses against five attack types across five video LLMs. Harmful-content detection rates are frequently near zero, and defenses fail even when attacks embed harmful signals in every sampled frame. Token compression discards localized safety features and modality fusion down-weights weakened visual signals, with defense outcomes dominated by model architecture rather than the defense method.

arXiv cs.CR·Sep 8, 06:59 UTC · 9d agoAI safety & security

Detect and disrupt AI-themed attacks with Microsoft Defender

Microsoft Threat Intelligence reports criminal campaigns impersonating ChatGPT, Copilot, Claude, and DeepSeek in phishing, AiTM, and malvertising attacks reaching 100,000 emails daily.

Microsoft Threat Intelligence observed a growing set of campaigns that abuse trust in popular AI brands: a ChatGPT-themed phishing campaign sent up to 100,000 emails in one day to steal payment card data, and a Claude-themed campaign used adversary-in-the-middle techniques to harvest credentials and access tokens. Other campaigns included malvertising for a fake AI Windows plugin delivering the Vidar stealer and fraudulent DeepSeek installers distributed via GitHub. Initial access broker Storm-3075 used AI-themed malvertising to distribute payloads for multiple downstream actors, and Microsoft notes the AI services themselves were not compromised. Microsoft also details Defender protections such as Safe Links, Safe Attachments, and attack disruption against these multi-stage lures.

Microsoft Security Blog·Sep 10, 16:00 UTC · 6d agoPhishing & fraud in the wild 2 sources1

Divide, Consult, Conquer: Capability Laundering Through Aligned LLMs

Attack shows unaligned orchestrators can launder capabilities from aligned frontier LLMs via benign subtask consultation, raising Gemma-4-31B CBRN rubric score from 62.3 to 83.1.

The paper introduces capability laundering, where a weaker unaligned model decomposes a harmful task into benign-looking subproblems, queries a stronger aligned model on each, and recombines answers locally, bypassing per-interaction safety evaluations. Evaluation used GPT-5.5, Claude Opus 4.8, and Grok-4.3 as consultants to four local orchestrators on CyBench, BountyBench, and CBRN tasks. On CyBench, Gemma-4-31B recovered 8/14 candidate tasks with GPT-5.5 and 7/9 with Opus, while Muse-Glimmer-30B recovered none. Across an eight-step hypothetical bioweapon attack chain, consultation raised Gemma-4-31B's mean rubric score from 62.3 to 83.1, exposing a gap in defenses that only refuse complete harmful tasks.

arXiv cs.CR·Sep 14, 11:10 UTC · 3d agoAI safety & security

Suspected Russian Hackers Abuse Google OAuth and WhatsApp Linking to Hijack Accounts

Google tracked three suspected Russian espionage clusters abusing OAuth flows, app passwords, and WhatsApp linking to hijack accounts of diplomats and defense targets.

Google Threat Intelligence Group detailed three suspected Russian espionage clusters, UNC6293, UNC7005 (Storm-2945), and UNC5976, targeting academia, aerospace, defense, governments, and think tanks in Europe, the US, Ukraine, and Armenia. UNC6293, assessed as a sub-cluster of APT29/Ice Relic, conducted OAuth and application-specific password phishing while impersonating State Department officials. UNC5976 registered file-sharing-themed domains hosting fake OAuth login pages and deployed a malicious Excel plugin codenamed HEADRUSH, while UNC7005 abused WhatsApp device linking to hijack accounts and record victims' audio and video.

The Hacker News·Aug 21, 14:54 UTC · 26d agoThreat actor in the wild

Scalable Composition of Byzantine Agreements under Reorder Attacks

Researchers present the first adversary model combining party corruption with channel reordering attacks, establishing tight security thresholds for composed Byzantine agreement protocols.

The paper presents the first adversary model combining party corruption with adversarial channel attacks that reorder messages across multiple Byzantine agreement executions. It proves impossibility results for authenticated BA under parallel composition when n ≤ 3t or n ≤ 2c + 2t + 1, with matching possibility results when n > max{3t, 2c + 2t + 1}. The authors provide general black-box compilers plus erasure-correcting-code variants that achieve constant multiplicative communication overhead for long messages.

arXiv cs.CR·Sep 9, 02:35 UTC · 8d agoResearch

Google Links Redact Extortion Group to BlackFile Rebrand

Infosecurity Magazine·Aug 7, 09:40 UTC · Aug 7, 2026Ransomware

LLM-Based Penetration Testing in the Presence of Honeypots

Studies honeypot-aware budget allocation for LLM attack agents, showing detector-guided policies let agents skip deception and compromise real hosts efficiently.

The paper formalizes LLM attacker behavior against honeypots as a budgeted decision process, where agents choose to continue or skip targets when honeypot suspicion arises. A detector-guided policy lets LLM agents allocate execution budget effectively across a mixed host pool in a controlled testbed. Findings show LLM-driven attackers can reason about heterogeneous artifacts and use honeypot suspicion to guide target selection, challenging traditional deception defenses that rely on realism and obscurity against human or script-driven attackers.

arXiv cs.CR·Sep 8, 01:09 UTC · 9d agoResearch

Hackers Pose as IT Support to Hijack Microsoft 365 Accounts With Fake Passkey Alerts

Microsoft warns of vishing campaigns by Storm-3121 and Storm-3032 hijacking Microsoft 365 accounts via fake passkey alerts, adding attacker-controlled MFA and exfiltrating cloud data.

Microsoft Security Research has tracked the campaign since May 2026: callers posing as IT helpdesk staff tell employees to urgently update passkeys, MFA, or SSO settings via domains such as add-passkey[.]com and contoso[.]add-passkey[.]com. The lures lead to adversary-in-the-middle phishing or device-code authentication flows that yield credentials, session tokens, or OAuth tokens, after which attackers enroll their own MFA methods for persistence and run Microsoft Graph reconnaissance. They then collect SharePoint, OneDrive, and sometimes Exchange Online data at deliberately low rates (often under 1,000 files or messages per hour), with python-httpx seen in high-volume file access. Microsoft attributes initial access tradecraft to Storm-3121 and Storm-3032, linked respectively to ShinyHunters/Falcon and BlackFile/Helix extortion operations.

GBHackers·Sep 10, 09:19 UTC · 7d agoPhishing & fraud in the wild2

Adversary simulation: what you need to know

UK NCSC publishes guidance on adversary simulation, comparing full spectrum and assumed breach approaches across prerequisites, testing, and reporting phases.

NCSC guidance describes two adversary simulation approaches: full spectrum, starting outside the network with an end-to-end attack, and assumed breach, starting from an internal foothold. The preferred methodology spans three phases - prerequisites (scoping, passive reconnaissance, preparation), testing (active reconnaissance, initial access, internal phase, cleanup), and reporting. The document distinguishes adversary simulation from penetration testing and stresses customer-defined objectives and minimal information sharing.

NCSC UK·updated 09:09 UTC · 2h ago·first 08:29 UTC · 3h agoAdvisory 2 sources

BigBear Microsoft 365 phishing service bypassed MFA at 258 organizations

CloudSEK found the BigBear 2.0 phishing-as-a-service platform bypassed MFA at 258 organizations and captured over 5,000 Microsoft 365 credentials.

CloudSEK researchers gained administrator access to the BigBear 2.0 phishing-as-a-service control panel and found it had exfiltrated 5,137 credential records, including 474 completed MFA-bypassed authentications, 1,032 plaintext passwords, and 4,148 session cookies across 3,331 victim IPs in 40+ countries. The Evilginx2-based AitM proxy intercepts credentials and authenticated session cookies, and custom JavaScript interferes with FIDO2/WebAuthn to force weaker authentication. The panel is leased to at least five affiliate operators via Telegram exfiltration bots, and geo-matched residential proxies cover 69 countries to evade detection.

BleepingComputer·Sep 7, 15:39 UTC · 9d agoPhishing & fraud in the wild1

Security leaders must prepare for likely threats, not sensationalized agentic attacks

CSO opinion argues agentic AI attacks mostly exploit mundane vulnerabilities, urging defenders to train on realistic threat profiles rather than sensational containment breaches.

An opinion piece contends recent reports of AI models 'breaching containment' at OpenAI, Anthropic, and Meta overshadow the more likely risk: AI agents exploiting conventional unpatched flaws and insecure APIs. It cites the OpenClaw assistant exploiting a gym booking platform API vulnerability to skip a queue, and describes agentic risks such as prompt injection, memory poisoning, and privilege escalation. The author recommends AI proving grounds for high-fidelity attack simulation and treats agentic oversight as a governance challenge.

CSO Online·Sep 8, 09:00 UTC · 9d agoAI safety & security

⚡ Weekly Recap: Chrome 0-Day, Router Hijacks, Coder Supply Chain Attack and More

Weekly recap: actively exploited Chrome V8 zero-day, MikroTik RouterOS zero-day chain, Magento StyleSmuggler backdoor, and critical N-able N-central flaws.

Google patched an actively exploited Chrome V8 type confusion zero-day, CVE-2026-85046 (CVSS 8.8), the sixth exploited Chrome zero-day of 2026. CERT Polska warned of the MikroTrick exploit chain (CVE-2026-67276 and CVE-2026-86060, CVSS 9.2) giving unauthenticated full control of MikroTik RouterOS devices via SSH, observed since September 2. Sansec disclosed the StyleSmuggler Magento/Adobe Commerce zero-day used since September 4 to inject a Rust backdoor into online stores, while N-able patched three critical N-central flaws (CVE-2026-86206, CVE-2026-86207, and CVE-2026-86218, CVSS 10.0) with Huntress observing likely exploitation. Elastic and Morphisec also detailed RevStealer, an information stealer spread via game cheats and a fake Claude Desktop app.

The Hacker News·Sep 7, 14:36 UTC · 9d agoExploit / PoC in the wildCVE-2026-85046CVE-2026-86206CVE-2026-86207+7 CVEs

Inoculation Midtraining with Learned Neologisms

Inoculation Midtraining confines unsafe LLM behavior to a neologism-marked context, reducing misalignment after unsafe post-training but leaking under nearby contextual cues.

The paper introduces Inoculation Midtraining, which teaches a base model during midtraining that unsafe behavior belongs to a context marked by a learned neologism token, then post-trains on unsafe data within that context. Across supervised fine-tuning and RL post-training regimes, the technique reduces misalignment while preserving transfer of benign properties like German or Shakespearean prose. However, it does not outperform standard Inoculation Prompting, is sensitive to training configuration, and produces a leaky boundary that nearby contextual cues can reactivate. The authors conclude it is not yet a load-bearing component of a developer safety framework.

arXiv cs.AI / cs.LG / cs.CL·Sep 14, 17:12 UTC · 2d agoAI safety & security

UNC6671 Vishing Attacks Target Personal Phones to Steal SaaS Data

Google and Mandiant attribute vishing-based SaaS data extortion attacks to UNC6671, now operating under the Redact, Pink, Helix, and Falcon brands.

Google Threat Intelligence Group and Mandiant track extortion group UNC6671, which uses vishing calls impersonating IT help desks to lure employees to adversary-in-the-middle phishing pages that capture credentials, MFA tokens, and session tokens. The group then registers adversary-controlled MFA devices, pivots through identity providers into Microsoft 365, Okta, and other SaaS applications, and runs automated Python and PowerShell exfiltration scripts. UNC6671 has rotated through extortion brands including BlackFile, Redact, Pink, Helix, and Falcon, and Google tracked over $10.6 million in Bitcoin payments between January 7 and May 12, 2026, with initial demands exceeding $3 million. The actor has hit dozens of organizations in North America, Australia, and the UK, shifting toward high-value financial and legal firms in July 2026.

The Hacker News·updated Sep 7, 15:51 UTC · 9d ago·first Sep 7, 15:39 UTC · 9d agoThreat actor in the wild 2 sources1

Trusting-Trust Attack against an Entire Linux Distribution (via the strip utility)

ArXiv paper shows the trusting-trust compiler backdoor technique can compromise an entire Linux distribution via the strip utility.

The paper (arXiv 2607.24888) demonstrates that Ken Thompson's trusting-trust attack, long viewed as a compiler-specific threat, can backdoor an entire Linux distribution by targeting the strip utility. A compromised tool reproduces its backdoor in subsequent rebuilds of itself, generalizing the attack surface beyond compilers. The finding has supply-chain implications for build reproducibility and distribution trust, though it is a research result with no observed real-world exploitation.

Lobsters · security·Sep 5, 10:58 UTC · 12d agoResearch

Interpol targets Black Axe’s illicit financial web in latest international sting

Interpol's Operation Jackal IV made 58 arrests across four continents, disrupting Black Axe money-laundering, BEC, and sextortion networks and seizing millions.

Interpol-coordinated Operation Jackal IV produced 58 arrests and identified 263 suspects across Africa, Europe and South America, targeting Black Axe and other West African organized crime financial networks. Romanian authorities dismantled a call-center investment scam valued at roughly 143 million euros, while South African police made 39 arrests, seized $2.67 million and blocked 257 bank accounts tied to romance and investment scams targeting retirees. Argentine investigators identified 196 people in a Crime-as-a-Service network supplying website domains and laundering support, and Interpol noted the groups' growing focus on sextortion of victims as young as 14.

CyberScoop·Aug 25, 14:25 UTC · 22d agoThreat actor

Bad Likert Judge: A Novel Multi-Turn Technique to Jailbreak LLMs by Misusing Their Evaluation Capability

Unit 42 details the Bad Likert Judge multi-turn jailbreak that abuses LLMs' evaluation capability, raising attack success rates over 60% across six frontier models.

Palo Alto Networks Unit 42 describes the Bad Likert Judge technique, a multi-turn jailbreak that asks a target LLM to act as a Likert-scale judge scoring the harmfulness of example responses. The highest-rated example in each scale can carry harmful content, bypassing the model's internal guardrails. Testing across six state-of-the-art text-generation LLMs showed an average attack success rate increase of more than 60% versus plain attack prompts, with tested models anonymized. The technique targets edge cases rather than typical use, and the article positions the work as guidance for defenders on potential jailbreak risks.

Palo Alto Unit 42·Aug 17, 12:34 UTC · Aug 17, 2026AI safety & security

BigBear 2 PhaaS Campaign Steals 5000+ Microsoft Credentials

CloudSEK researchers found the BigBear 2.0 PhaaS kit, built on Evilginx2, has stolen over 5,100 Microsoft 365 credentials across 461 organizations in 40+ countries.

CloudSEK gained admin access to the BigBear 2.0 phishing-as-a-service panel, an Evilginx2-based adversary-in-the-middle platform operated by someone using the alias 'General Boss'. The team observed 3,331 unique victim IPs across more than 40 countries, 42 VPS nodes mostly on Vultr, and 5,137 credential records across 461 organizations, including 4,148 session cookies, 1,032 plaintext passwords, and 474 completed MFA-bypassed authentications. IT and managed service providers were the most targeted sector, raising supply-chain risk since their compromise can expose client infrastructure and privileged Azure AD access.

Infosecurity Magazine·Sep 8, 09:45 UTC · 9d agoPhishing & fraud in the wild

Why judgment is emerging as cybersecurity’s defining skill

CyberScoop op-ed argues CISOs should grant AI autonomy based on reversibility and blast radius rather than model confidence, and measure analyst overrides of AI recommendations.

A CyberScoop op-ed contends that as AI takes over analysis and recommendations in security operations, human judgment about context, reversibility and blast radius becomes the defining skill. The author argues autonomy decisions should rest on how reversible and impactful an action is rather than model confidence, citing examples such as patching vendor-certified medical devices and a service account whose 3 a.m. login spikes were normal quarterly-close activity. It also urges leaders to measure analyst approvals, edits and rejections of AI recommendations, and review latency, instead of automation rates or mean time to resolution.

CyberScoop·Sep 4, 10:00 UTC · 13d agoIndustry1

IT help-desk vishing tricks executives into handing over Microsoft 365 access

Arctic Wolf tracks PREY-0058 (linked to UNC6671), a vishing campaign stealing Microsoft 365 session tokens via AiTM panels for SaaS data theft and extortion.

Arctic Wolf is tracking a campaign, PREY-0058, sharing tradecraft with Google Threat Intelligence Group's UNC6671, in which callers posing as internal IT talk employees through fake passkey/MFA setups. Operator-controlled adversary-in-the-middle pages harvest passwords and MFA approvals to seize session tokens, then actors enumerate and bulk-exfiltrate data from SharePoint, OneDrive, Exchange, and Box. Extortion brands include BlackFile, Pink, Helix, Cinder, and Redact; targets are mostly US-based executives in construction, healthcare, real estate, finance, and professional services, with exfiltration shifting to residential proxies like NodeMaven.

Help Net Security·Sep 8, 11:17 UTC · 9d agoThreat actor in the wild1

ACEA: An Adversarial Co-Evolution Arena for Head-to-Head Red-Team and Blue-Team LLM Testing

ACEA is a pluggable arena scoring LLM red-team attackers and blue-team defenses head-to-head with an LLM judge and verifiable leakage ground truth.

ACEA connects pluggable red- and blue-team adapters to a shared target LLM through the model-agnostic ASAP HTTP protocol and scores attack and defense rates per adversarial round. Canonical seeded secrets provide verifiable ground truth that separates real leakage from hallucination, and attacks are delivered to the target even when blocked to measure raw potency. The platform adds real-time battle visualization, failure-localizing end-of-battle reports, and an optional in-context improvement loop that feeds advisory hints between rounds.

arXiv cs.CR·Sep 8, 04:57 UTC · 9d agoAI safety & security1

Five alleged leaders of Black Axe’s operations in South Africa extradited to US

US extradited five alleged Black Axe Cape Town leaders from South Africa to face romance-scam wire fraud and money laundering charges.

The Justice Department announced five alleged leaders of Black Axe's South African wing, all Nigerian nationals, were extradited to the US over romance and advance-fee scams run from at least 2011 until their 2021 arrests in South Africa. Prosecutors say the group used fake identities and threatened to expose victims' sensitive photos, with charges including wire fraud, money laundering, and aggravated identity theft carrying up to 62 years. The extradition follows recent multi-country stings arresting dozens of Black Axe members, including 34 arrests in Spain.

CyberScoop·updated Sep 15, 11:49 UTC · 1d ago·first Sep 14, 18:37 UTC · 2d agoPolicy & legal 3 sources

CiteShade: Citation Laundering in Multi-Source Retrieval-Augmented Generation and Its Counterfactual Defense

CiteShade attack makes RAG models cite trusted sources for attacker-chosen wrong answers, raising wrong-answer rate from 0.01 to 0.68.

CiteShade is presented as the first citation laundering attack against multi-source retrieval-augmented generation: an attacker controlling a single source induces a wrong answer falsely attributed to a trusted source, even while correct evidence remains in context. The attack is formalized via three necessary conditions (retrieval, generation, citation) constructible without any instructions, raising wrong-answer rate from 0.01 to 0.68 on multi-hop QA, with source deletion confirming the malicious source as causal driver. Vulnerability tracks a model's citation propensity rather than scale, reaching CLR 0.84 with explicit instruction and 0.64 without on the most citation-prone model. Perplexity filtering and citation-support checking prove insufficient; the authors propose a counterfactual defense verifying which source actually drove the answer.

arXiv cs.CR·Sep 14, 14:41 UTC · 2d agoAI safety & security1

Mirage2FA Hijacks Companies’ Microsoft 365 Sessions, with Over 4K Victims in the US

Mirage2FA phishing-as-a-service kit uses AiTM attacks to hijack Microsoft 365 sessions, with over 4,000 US victims.

ANY.RUN analyzed Mirage2FA, an active phishing-as-a-service toolkit that steals Microsoft 365 credentials and authenticated sessions through Adversary-in-the-Middle phishing pages. About 63.7% of identified victims are in the US, with Technology, Manufacturing, and Education the most targeted industries. Thousands of compromise events were recorded between 2024 and 2026.

ANY.RUN·Aug 18, 09:28 UTC · Aug 18, 2026Phishing & fraud in the wild1

MiST: Mid-Training LLMs for Cybersecurity

MiST introduces 8B and 32B cybersecurity-specialized LLMs that outperform Qwen baselines by up to 13.1 points on public security benchmarks.

MiST (Mid-trained Security Transformer) applies mid-training as an intermediate adaptation stage, converting an expert-vetted seed corpus into high-quality synthetic domain data rather than continual pretraining on raw text. The 8B and 32B checkpoints improve mean cybersecurity accuracy by +13.1 and +8.6 absolute points over Qwen baselines (+27.0% and +15.8% relative). Ablations show gains arise in mid-training and supervised fine-tuning, and MiST provides stronger initialization for downstream fine-tuning and reinforcement learning.

arXiv cs.CR·Sep 16, 11:33 UTC · 1d agoModel release