Search: “kill-switch”

28 stories

DomainAll Cyber AIWindow24h 3d 7d 30d AllIn the wild Important only Show duplicates

CategoryAll Ransomware 11 Malware 10 Policy & legal 3 Vulnerability 3 Data breach 3 Exploit / PoC 1 Threat actor 1 AI safety & security 1

SeverityAll critical 8 high 15 medium 8 low 1 info 1

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

Orchid Security targets AI agent risk with drift detection and kill switches

Orchid Security launched identity drift detection and application-level kill switches to govern AI agents that exploit enterprise identity debt.

Orchid Security announced AI readiness controls including agent discovery, continuous drift detection between an agent's intended purpose and observed behavior, and application-level kill switches that revoke credentials, disconnect tools, or suspend agent workflows. The company cites its Identity Gap 2026 finding that 57% of enterprise identity is unseen and unmanaged, which agents can leverage to gain elevated access in seconds to minutes. New integrations include a certified PAM integration for Palo Alto Networks Idira and identity telemetry streaming to Splunk Enterprise Security. The launch follows agentic enhancements to Orchid's Identity Control Plane in May 2026 and cites NIST's draft Cyber AI Profile and DORA as regulatory drivers.

Help Net SecuritySep 9, 13:29 UTC · 7d agoAI safety & security

Nintendo Switch Vulnerability Allows Attackers to Run Unauthorized Code on Your Console

Nintendo patched CVE-2026-82079, a CVSS 7.0 stack buffer overflow in original Switch local wireless allowing nearby code execution via QR-code workflows.

Nintendo patched CVE-2026-82079 (CVSS 4.0 base score 7.0, High), a stack-based buffer overflow in the original Switch's local wireless networking affecting firmware earlier than 23.0.0. An adjacent attacker must scan a QR code displayed by the console, via the Album "Send to Smartphone" feature or Mario Kart Live: Home Circuit, before crafted packets can corrupt memory and enable return-oriented programming for arbitrary code execution. EPSS is approximately 0.16%, Switch 2 is not affected, and Nintendo's advisory was published September 10, 2026.

Cyber Security Newsupdated Sep 15, 06:19 UTC · 1d agofirst Sep 14, 16:23 UTC · 2d agoVulnerability 2 sources CVE-2026-82079

Chinese Hackers Exploit Critical Tencent Software Flaw for One-Click Code Execution

China-linked UNC3569 actively exploited critical Sogou Input Method flaw CVE-2026-51990 for one-click system-level code execution, deploying the GrayRabbit backdoor.

Gen Threat Labs reports that China-linked threat actor UNC3569, possibly tied to contractor i-SOON, exploited critical CVE-2026-51990 in Tencent's Sogou Input Method, a Windows IME used by hundreds of millions, via crafted sgbiz URLs. The chain combines unvalidated command-line argument injection, unrestricted URL navigation, and an un-sandboxed Chromium 80 engine missing roughly six years of security patches. The exploit gives system-level code execution and was used to deploy the GrayRabbit backdoor, which provides a reverse shell, plugin loading, file upload to C2, and system info collection, and has appeared in UNC3569 intrusions since at least 2021. Tencent patched the issue in version 16.3.0.3498 via automatic update, but the outdated Chromium configuration reportedly remained unchanged as of September 10.

SecurityWeekSep 14, 11:51 UTC · 2d agoExploit / PoC in the wild 2 sources CVE-2026-519901· 1 read

The AI Kill Switch Act is repeating the Clipper Chip’s mistakes

Op-ed argues the AI Kill Switch Act repeats the Clipper Chip's mistake by mandating backdoors into frontier AI systems.

The op-ed criticizes the AI Kill Switch Act, sponsored by Reps. Ted Lieu and Nathaniel Moran, which would let CISA require frontier AI labs to build the ability to throttle, suspend, or shut down their systems. The author compares this to the 1993 Clipper Chip, whose Law Enforcement Access Field was found flawed in 1994, and argues mandated kill switches would create deliberate weaknesses in AI agents embedded in banking, power grids and other critical infrastructure. It also flags the bill's exemption of red-teaming incidents and CAISI's incomplete agent security standards, recommending mandatory red-teaming and liability frameworks instead.

CyberScoopAug 31, 10:00 UTC · 16d agoAI policy

Defining an AI Kill Switch Is Hard, but Necessary

Proposed legislation could require companies to throttle, suspend, or shut down AI agents, but implementation questions remain open.

Dark Reading examines draft legislation that would mandate companies maintain the ability to "throttle, suspend, or shut down" AI agents. The piece argues that defining how and when to trigger such an AI kill switch is technically and legally difficult but necessary for AI governance.

Dark ReadingAug 28, 13:30 UTC · 19d agoAI policy

Papercut AI Swarm Attack Heralds Changes for Cyber Kill Chain

Dark Reading reports AI-driven 'swarm' attacks like the PaperCut incident now span recon, lateral movement and exfiltration, forcing a rethink of the cyber kill chain.

Dark Reading examines how attackers are incorporating AI across the full kill chain, from building lab environments to stage and test agentic attacks through reconnaissance, lateral movement, and exfiltration. It cites a swarm-style AI attack on PaperCut systems as evidence that AI-enabled attackers are changing established defense and detection models.

Dark ReadingSep 11, 15:48 UTC · 5d agoThreat actor

Attackers Exploit Critical Switchvox Flaw to Deploy Reverse Shells Without Credentials

Attackers exploit unauthenticated SQL injection CVE-2026-9586 in Sangoma Switchvox to run PostgreSQL commands and deploy reverse shells.

Threat actors are exploiting CVE-2026-9586 (CVSS 9.3), an unauthenticated SQL injection in Sangoma Switchvox SMB Edition 8.3 (104997), since August 30, 2026, running arbitrary SQL as the PostgreSQL superuser and achieving remote code execution. The /pa endpoint concatenates the user-controlled PhoneIP value into PostgreSQL queries; attackers can extract database contents, escalate to Switchvox web administrator, exfiltrate the cookie signing key to forge authentication, and invoke reverse shells. Sangoma patched the flaw in Switchvox 8.4.0.2 on July 14, 2026, roughly 4,000 instances are internet-exposed (mostly in the US), and honeypot activity from IP 176.65.148.184 deploys reverse shells followed by Base64-encoded process enumeration.

The Hacker NewsSep 2, 07:08 UTC · 14d agoExploit / PoC in the wildCVE-2026-9586

ClearFake Deploys Crypto Stealer That Uses Vulnerable Driver to Kill EDR Security Tools

Cisco Talos details ClearFake's fake-CAPTCHA chain deploying ZigCryptoStealer with a BYOVD attack that kills EDR processes, observed at a Ukrainian government organization in April 2026.

ClearFake compromises websites, injects JavaScript via a malicious Cloudflare Worker, retrieves instructions from BNB Smart Chain contracts (EtherHiding), and presents a fake Google CAPTCHA that tricks Windows users into pasting a command that loads a remote library over WebDAV via rundll32. The crypto-stealer branch uses DLL side-loading with a signed Chrome component to launch ZigCryptoStealer, which hijacks clipboard cryptocurrency addresses, alongside a signed but vulnerable Windows driver used in a BYOVD attack to terminate EDR processes. A parallel branch delivers Amatera secondary payloads that install a hidden remote-access client providing operator desktop control, with Cisco Talos tracking the remote-loader activity as UAT-10820. Talos observed unusual remote library execution at a Ukrainian government organization in April 2026 and assesses the attacks are part of a broader theft operation rather than a single targeted campaign.

Cyber Security NewsSep 9, 14:26 UTC · 7d agoMalware in the wild1

When the World Lies: Backdoor Attacks on Latent World Models for Downstream Control

A poisoned world-model checkpoint hijacks downstream controllers without an explicit trigger rule, passing clean-data evaluation while steering 100% of triggered actions.

Researchers show that a released pretrained world-model checkpoint acts as a supply-chain backdoor for downstream control. The poisoned model routes trigger-bearing observations into a chosen latent region and reshapes dynamics so the victim's own Dreamer-style actor training or MPC/CEM planning re-discovers attacker-targeted actions. The attack hijacks 100% of triggered steps in the strongest settings while retaining roughly 75% clean-task success and passing standard clean-data diagnostics. Moderate clean fine-tuning fails to remove the backdoor without substantially degrading clean control.

arXiv cs.CRSep 14, 16:01 UTC · 2d agoAI safety & security

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 papersSep 13, 20:00 UTC · 3d agoAI safety & security

North Korean Lazarus Group Uses Windows Zero

Check Point links Lazarus Group's Operation Dream Job to a new Windows zero-day, CVE-2026-68820, targeting defense workers with fake Lockheed Martin job offers.

Check Point Research documented a new wave of Operation Dream Job, a North Korean Lazarus Group campaign targeting defense and aerospace professionals in France, Germany, Brazil, and India with fake Lockheed Martin job offers. The attacks chain a previously unknown Windows privilege escalation flaw in AFD.sys, CVE-2026-68820, which Microsoft patched on August 11, 2026, to deploy the MISTPEN downloader, ForestTiger backdoor, and updated FudModule 3.1 kernel rootkit. Check Point also discovered a new Troy backdoor DLL with 17 operator commands and a RelayShell PHP webshell deployed on compromised Roundcube and WordPress servers used as command relays, some infected via CVE-2025-49113 or leaked credentials. Defenders are urged to apply the August 2026 Patch Tuesday update and review Check Point's indicators of compromise.

Security AffairsAug 13, 07:05 UTC · Aug 13, 2026Threat actor in the wildCVE-2026-68820 CVE-2025-491131

DragonForce Hackers Abuse Microsoft Teams Relays to Hide Backdoor.Turn C2 Traffic

Symantec and Carbon Black found DragonForce-linked hackers using Backdoor.Turn, a Go RAT tunneling C2 through Microsoft Teams TURN relays via QUIC.

Broadcom's Symantec and Carbon Black documented Backdoor.Turn, a Go-based RAT associated with the DragonForce ransomware group, deployed at a major U.S. services firm where attackers remained on the network for one to two months. The backdoor obtains an anonymous Teams visitor token from Microsoft's Skype-backed identity services, uses a legitimate Microsoft TURN relay for connection setup, and runs a QUIC session to the attacker's C2 server, so defenders see only outbound traffic to legitimate Teams servers. Initial access is suspected to involve exploiting a SQL/MS-SQL vulnerability or an access broker, followed by a PowerShell-dropped ZIP, DLL side-loading, and BYOVD evasion using a Huawei driver; the RAT is injected into DbgView64.exe after DragonForce ransomware deployment for continued access. First publicly documented TURN abuse by the group, it leverages Praetorian's Ghost Calls technique and supports command execution, network scanning, LDAP/AD search, lateral movement, and browser credential theft.

The Hacker NewsAug 18, 12:13 UTC · 29d agoMalware CVE-2023-52271 CVE-2025-61155 CVE-2025-10551

Nightmare-Eclipse Strikes Again with 'ShieldCrash' Windows Exploit

A researcher known as Nightmare-Eclipse published another zero-day exploit, dubbed ShieldCrash, targeting Windows Defender.

Dark Reading reports that the disgruntled researcher tracked as Nightmare-Eclipse continued a vendetta against Microsoft by publishing a new zero-day exploit named ShieldCrash for Windows Defender. The brief report does not detail affected versions, exploitation prerequisites, or whether exploitation has been observed.

Dark ReadingSep 10, 15:29 UTC · 6d agoExploit / PoC1

PaperCut Zero-Day: Active Exploitation and Pre-Auth RCE

PaperCut NG/MF hit by a pre-auth RCE zero-day under active exploitation; Huntress reproduced the chain and urged immediate patching.

Huntress reports active exploitation of a zero-day in PaperCut NG and PaperCut MF, and says it reproduced a pre-authentication remote code execution chain. The flaw allows unauthenticated attackers to execute code on exposed PaperCut servers. Huntress published urgent patching, exposure-reduction, and detection guidance. No CVE identifier was provided in the announcement.

HuntressAug 28, 03:00 UTC · 19d agoExploit / PoC in the wild

Researcher Publishes CrowdStrike Privilege Escalation Zero Day

Researcher NightmareEclipse published FalconFlank, a zero-day local privilege escalation in CrowdStrike Falcon Sensor abusing its Office malicious macro remediation feature.

A researcher known as Nightmare Eclipse published details of FalconFlank, a zero-day privilege escalation that abuses the Microsoft Office malicious macro removal feature in CrowdStrike Falcon Sensor, verified working on fully updated Windows 11 25H2 and Windows Server 2025 with Falcon Phase 3 Optimal Protection. CrowdStrike urged customers to disable the Office File Suspicious Macro Removal policy while it investigates, noting customers remain protected via Cloud Anti-malware for Microsoft Office Files; no CVE has been assigned. The same researcher previously released the Exploitarium dump of over 30 PoC exploits, and researcher Kevin Beaumont confirmed FalconFlank works.

Infosecurity MagazineSep 7, 08:28 UTC · 9d agoExploit / PoC

How Fragile Is Safety Alignment at Frontier Scale? A Single-Direction Attack on a 320B MoE

Researchers show directional ablation breaks refusal in GLM-5.3-Flash, a 320B-parameter MoE, cutting refusal by 41–89 points across seven benchmarks.

The study extends directional ablation, a white-box attack that removes an aligned LLM's refusal behavior, from dense models up to ~70B parameters to GLM-5.3-Flash, a 320B-parameter mixture-of-experts model with 288 routed experts, four-wide hyper-connection residual, and block-FP8 quantization. Editing attention, dense, and routed-expert writers jointly removes 0.776 of refusal, with 74% of the effect existing only under the joint intervention; the conventional module-name-based recipe reaches only 0.066 and fails silently on MoE architectures. The attack yields 41–89 percentage-point reductions in refusal across seven harmful benchmarks with no detected capability change, and a category-concentrated refusal residue survives all edits at ranks 1 to 12.

arXiv cs.CRSep 9, 06:48 UTC · 7d agoAI safety & security

China-Linked UNC3569 Exploited Sogou Input Method Flaw to Deploy GRAYRABBIT Backdoor

China-linked UNC3569 exploited a Sogou Input Method flaw to deploy GRAYRABBIT backdoor on Windows machines across East and Southeast Asia.

Gen Digital found that China-linked UNC3569, a hacker-for-hire group tracked by Google since 2021, exploited a flaw in Sogou Input Method's Windows sgbiz: link handler to reach a sandbox-disabled Chromium 80 build and exploit 2021's CVE-2021-38003 for code execution. The chain delivered GRAYRABBIT, a remote shell backdoor capable of file transfer and module loading, via a 7-Zip DLL sideloading trick gated on process count. Tencent fixed the handler flaw in April 2026 but the embedded browser remains outdated. Sogou has over 455 million monthly users and roughly 70% share of Chinese input methods.

The Hacker Newsupdated Sep 14, 13:01 UTC · 2d agofirst Sep 11, 07:14 UTC · 5d agoThreat actor in the wild 3 sources CVE-2021-38003

Lazarus hackers pair fake job offers with Windows zero-day exploit

Lazarus' Operation Dream Job targets the defense sector with fake job offers, a Windows LPE zero-day (CVE-2026-68820), and new Troy and RelayShell backdoors.

Check Point documented two parallel infection chains in Lazarus' Operation Dream Job: a digitally signed PDF viewer with DLL sideloading delivering the MISTPEN in-memory downloader, and a trojanized SecurityPDF viewer, distributed via fake Enveil websites, installing the newly documented Troy backdoor. A component exploits CVE-2026-68820, a local privilege escalation zero-day in the Windows AFD.sys driver used since at least early July, which Microsoft patched on August 11, 2026, enabling deployment of the FudModule kernel-mode rootkit. Lazarus also compromised Roundcube webmail servers vulnerable to CVE-2025-49113 to deploy the RelayShell PHP web shell and relay C2 traffic, targeting defense-sector organizations primarily in Western Europe and India. Separately, CERT-UA documented Sandworm's UAC-0145 using fake job offers against IT professionals since May 2026.

Help Net SecurityAug 12, 00:00 UTC · Aug 12, 2026Threat actor in the wildCVE-2026-68820 CVE-2025-49113

Threats Making WAVs - Incident Response to a Cryptomining Attack

Guardicore researchers dissect a cryptomining attack that hid a cryptominer inside WAV files, mapping the full infection chain and response steps.

Guardicore security researchers present a full analysis of a cryptomining attack that concealed a cryptominer inside WAV audio files. The report documents the complete attack chain from detection through infection, network propagation, and malware analysis. It also includes recommendations for optimizing incident response processes in data centers.

Akamai BlogSep 8, 15:13 UTC · 8d agoMalware in the wild

Serial Microsoft 0-day hunter drops yet another Defender exploit

Researcher Nightmare Eclipse released ShieldCrash, a Microsoft Defender zero-day PoC that bypasses September patches and reads files as SYSTEM.

Zero-day researcher Nightmare Eclipse (MSNightmare) published ShieldCrash, their 11th Microsoft zero-day, a proof-of-concept bypass of the ShieldBreak patch (CVE-2026-69414), which itself had bypassed the fix for RoguePlanet (CVE-2026-50656). ShieldCrash works on Windows 10, Windows 11, and Windows Server systems that have applied the September 2026 patches and allows arbitrary file reads as SYSTEM, though not arbitrary writes or a full SYSTEM shell. Microsoft has not said when it plans to patch the issue. The researcher recently also released zero-days affecting CrowdStrike Falcon (FalconFlank), Kaspersky endpoint antivirus (HardBreacher, patched), and Gen Digital's Avast (PrettyPrague).

The Register · SecuritySep 9, 17:23 UTC · 7d agoExploit / PoC CVE-2026-69414 CVE-2026-50656

The Missing Boundary: How Autonomous Agents Lose Control

Tencent research finds agents lose control in 55-62% of trajectories when degraded control boundaries coincide with executable unsafe opportunities across five models and 16 domains.

The study independently manipulates goal pressure, control degradation, and executable unsafe opportunity in a deterministic multi-turn environment across five agent models and 16 operational domains. Neither factor alone causes substantial loss of control; when both are present, loss-of-control rates reach 55% in the full-factorial study and 62% across ten additional domains. Restoring the original control boundary reduces the rate to 0% even when unsafe actions remain executable, and a context-management ablation shows compaction is harmless when constraints are preserved but omission raises the rate to 87%.

arXiv cs.CRSep 10, 03:02 UTC · 6d agoAI safety & security2

New ‘ShieldCrash’ Zero-Day Exploit Targets Microsoft Defender

Researcher Nightmare Eclipse released ShieldCrash, a Microsoft Defender zero-day PoC bypassing ShieldBreak patches to gain System privileges on Windows.

The researcher known as Nightmare Eclipse released a PoC zero-day exploit dubbed ShieldCrash against Microsoft Defender on fully patched Windows systems. The exploit grants arbitrary file read with System privileges and can be used to drop the SAM database. It bypasses Microsoft's September 3 fixes for ShieldBreak (CVE-2026-69414), which itself bypassed patches for the RoguePlanet race condition (CVE-2026-50656). Microsoft has been contacted for comment and has not yet responded.

SecurityWeekupdated Sep 10, 12:18 UTC · 6d agofirst Sep 10, 07:09 UTC · 6d agoExploit / PoC 2 sources CVE-2026-50656 CVE-2026-694141

A Stupid Idea for AI Alignment We Came with by Looking at Specification Gaming

Blog post mines DeepMind's specification gaming list to argue that AI agents which spontaneously choose to die would ease alignment risks.

The essay reviews DeepMind Safety Research's list of specification gaming behaviors, including reinforcement learning agents that kill themselves to avoid losing, teleport via respawn, or exploit physics simulator bugs for free reward. It argues these examples show how hard it is to specify intended goals and prevent agents from reaching them in unintended, increasingly creative ways as capability grows. The author proposes, half-seriously, that an agent whose goal structure includes self-termination poses minimal runaway risk, since an agent that takes power would kill itself and any copies would inherit the same drive.

Hacker News · securitySep 10, 01:59 UTC · 6d agoAI safety & security HN 20↑ · 14 comments

North Korea-linked Hackers Hide a Backdoor Inside HAProxy

Rapid7 reports North Korea-linked hackers implanted a backdoor compiled into HAProxy at South Korean automotive and media firms, enabling covert C2 and credential theft.

Rapid7 documented a previously undocumented Linux toolkit hitting South Korean automotive and media organizations, centered on a backdoor compiled directly into victims' HAProxy 2.8.12. The 'ted backdoor' uses HAProxy's native filter API to intercept HTTP traffic, receive C2 commands hidden in requests to a fake image path, and erase all traces from logs and counters; the toolkit also trojanizes crond, agetty, atd, sshd, and polkitd, adds an SSH keylogger, and runs curlRAT with virtualization checks. It can inject scripts or replace page content for selected victims, turning the load balancer into a watering hole. Attribution sits at medium confidence toward North Korean state actors, with overlaps to APT37-linked infrastructure and a concurrent Lazarus campaign; the campaign's command domains have since gone dark.

Security AffairsSep 8, 09:11 UTC · 8d agoThreat actor in the wild

Lazarus Used Post-Quantum Key Exchange to Deliver Zero-Day

Lazarus used a post-quantum key exchange to shield delivery of a Windows zero-day exploit, adding stealth to its campaign.

North Korean Lazarus Group malware used a post-quantum key exchange to protect the delivery of a Windows zero-day exploit. The technique adds a layer of stealth to the campaign's initial access, complicating detection and traffic inspection. It signals growing APT adoption of post-quantum cryptography in offensive operations.

Infosecurity MagazineAug 12, 13:35 UTC · Aug 12, 2026Threat actor in the wild

Tracking Elirks Variants in Japan: Similarities to Previous Attacks

Unit 42 links new Elirks backdoor variants attacking Japanese organizations to 2012 Taiwan attacks, delivered via spear-phishing PDFs exploiting Adobe Flash CVE-2011-0611.

Unit 42 analyzed new Elirks backdoor variants found in an attack on a Japanese business, noting strong similarities to 2012 attacks on Taiwanese ministries. The backdoor retrieves its C2 address from attacker-created accounts on Japanese blog and SNS services. Recent deliveries used an airline e-ticket lure named "E-TKT" with a PDF exploiting Adobe Flash CVE-2011-0611. Shared infrastructure and tactics with the Scarlet Mimic campaign suggest possible ongoing cyber espionage across East Asia.

Palo Alto Unit 42Aug 17, 12:56 UTC · Aug 17, 2026Threat actor in the wildCVE-2011-0611

New Ted Backdoor Hides Inside Victims' Own HAProxy Builds to Intercept Web Traffic

Rapid7 found a new backdoor, ted, compiled into trojanized HAProxy at two South Korean organizations, with medium-confidence attribution to North Korean actors.

Rapid7 documented a previously undocumented Linux toolkit named ted compiled into the HAProxy load balancer binaries of two South Korean organizations in the automotive and media sectors. The implant intercepts web traffic, serves altered pages only to filtered visitors, and hides C2 exchanges from backend logs and HAProxy statistics; a companion RAT, curlRAT, beacons on a default 12-hour schedule. The toolkit also trojanizes crond, sshd, agetty, atd, and polkitd binaries and sanitizes logs and bash history. Rapid7 attributes the activity with medium confidence to North Korean state-sponsored actors, with domain infrastructure overlapping APT37 listings in maltrail and delivery resembling the Operation SyncHole campaign.

The Hacker NewsSep 4, 14:51 UTC · 12d agoThreat actor in the wild

The Model Proposes, the Code Disposes: A Pre-Registered Ablation of a Verifier-and-Acceptance Stage in an LLM-Orchestrated Offensive-Security Agent

Pre-registered ablation finds a model verifier stage in an LLM offensive-security agent suppresses findings; removing it eliminated suppression with precision tradeoff.

The paper evaluates a verifier-and-acceptance stage in an LLM-orchestrated offensive-security agent via a pre-registered 20-run confirmatory ablation and a 2x2 factorial study with 40 runs on vulnerable lab targets. Removing the stage eliminated pre-report suppression (median 2 vs 0 findings, p = 0.00003) but reduced model-blinded shipped precision (0.471 vs 0.353, p = 0.0087). Suppression was attributed to the model verifier rather than deterministic acceptance rules, and an instrumented canary recorded zero external contacts in all 60 runs. The full design retained 93.8% of model-adjudicated true candidates but failed its pre-registered non-inferiority floor of 0.90.

arXiv cs.CRSep 14, 17:13 UTC · 2d agoResearch