Search: “race condition”

40 stories

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

CategoryAll Vulnerability 246 Exploit / PoC 116 Malware 27 Policy & legal 26 Ransomware 21 Data breach 18 Threat actor 13 Advisory 13 Industry 12 AI safety & security 10 AI industry 4 Phishing & fraud 4 Research 4 AI tools & infra 2

SeverityAll critical 254 high 122 medium 87 low 38 info 17

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

ZDI-26-568: Linux Kernel Net Scheduler Race Condition Local Privilege Escalation Vulnerability

ZDI disclosed a race condition (CVSS 7.5) in the Linux kernel net scheduler enabling local privilege escalation; no CVE assigned in the advisory text.

ZDI advisory ZDI-26-568 describes a race condition in the Linux kernel's net scheduler that allows local attackers to escalate privileges on affected installations. Exploitation requires the attacker to first execute high-privileged code on the target system. ZDI assigned a CVSS rating of 7.5; no CVE identifier is listed in the advisory text.

ZDI Published AdvisoriesAug 13, 05:00 UTC · Aug 13, 2026Vulnerability

ZDI-26-574: Linux Kernel Net Scheduler Connection Tracking Race Condition Local Privilege Escalation Vulnerability

A race condition (CVE-2026-46319, CVSS 7.5) in the Linux kernel net scheduler connection tracking allows local privilege escalation.

ZDI advisory ZDI-26-574 documents a race condition in the Linux kernel's net scheduler connection tracking component. Local attackers who can execute high-privileged code on a target can exploit the flaw to escalate privileges. ZDI rated the vulnerability 7.5 on CVSS and assigned CVE-2026-46319.

ZDI Published AdvisoriesAug 13, 05:00 UTC · Aug 13, 2026Vulnerability CVE-2026-46319

Testing race conditions with memory access tracing and stack-based delay injection

Google Project Zero released MAccConc, Linux kernel tooling that traces memory accesses to explore and test race condition interleavings.

A Google Project Zero researcher published MAccConc (Memory Access Concurrency), tooling for exploring possible interleavings of multithreaded test cases in the Linux kernel, available on GitHub. The tools use KCOV with ASAN outline-mode instrumentation to record per-access memory traces, enabling automatic testing of all A-B-A interleavings plus terminal and GUI explorers for manual analysis. The work targets confirming race condition candidates, building reliable regression tests, and enabling concurrency fuzzing, drawing on ideas from SKI and Ned Williamson's sockfuzzer.

Google Project ZeroSep 8, 07:00 UTC · 8d agoResearch1

ZDI-26-693: Linux Kernel ksmbd Share Configuration Race Condition Remote Code Execution Vulnerability

ZDI-26-693: authenticated race condition in Linux kernel ksmbd share configuration allows remote code execution on ksmbd-enabled systems; CVSS 8.5.

ZDI advisory ZDI-26-693 discloses a race condition in the Linux kernel's ksmbd share configuration that allows remote attackers to execute arbitrary code on affected installations. Exploitation requires authentication, and only systems with ksmbd enabled are vulnerable. ZDI assigned a CVSS rating of 8.5; no CVE is listed in the advisory text.

ZDI Published AdvisoriesSep 14, 05:00 UTC · 2d agoVulnerability1

ZDI-26-692: Linux Kernel eMPIA USB Device Driver Race Condition Code Execution Vulnerability

ZDI-26-692: race condition in Linux kernel eMPIA USB driver (CVE-2026-31583) lets physically present attackers execute code without authentication; CVSS 7.1.

ZDI advisory ZDI-26-692 discloses a race condition in the Linux kernel's eMPIA USB device driver that allows physically present attackers to execute arbitrary code on affected installations. Authentication is not required, but physical access to the target system is necessary. ZDI assigned a CVSS rating of 7.1 and CVE-2026-31583.

ZDI Published AdvisoriesSep 14, 05:00 UTC · 2d agoVulnerability CVE-2026-31583

ZDI-26-644: Oracle VirtualBox VMSVGA Race Condition Local Privilege Escalation Vulnerability

ZDI publishes ZDI-26-644 for CVE-2026-60155, a race condition local privilege escalation in Oracle VirtualBox VMSVGA, rated CVSS 7.5.

Zero Day Initiative published advisory ZDI-26-644 describing a race condition in Oracle VirtualBox's VMSVGA component. Local attackers who already execute high-privileged code on the guest system can escalate privileges on affected installations. ZDI rated the issue CVSS 7.5 and assigned CVE-2026-60155.

ZDI Published AdvisoriesSep 9, 05:00 UTC · 7d agoAdvisory 6 sources CVE-2026-60155

ZDI-26-695: Linux Kernel NFSv4 Server Race Condition Remote Code Execution Vulnerability

ZDI-26-695: Linux Kernel NFSv4 server race condition (CVE-2026-89688, CVSS 8.5) enables remote code execution on nfsd systems with authentication.

ZDI advisory ZDI-26-695 describes a race condition in the Linux Kernel NFSv4 server tracked as CVE-2026-89688 with a CVSS score of 8.5. Remote attackers can execute arbitrary code, but authentication is required and only systems with nfsd enabled are vulnerable. No in-the-wild exploitation is mentioned in the advisory.

ZDI Published AdvisoriesSep 14, 05:00 UTC · 2d agoVulnerability CVE-2026-896881

ZDI-26-702: Linux Kernel usbnet Driver Race Condition Privilege Escalation Vulnerability

ZDI discloses Linux Kernel usbnet driver race condition (CVE-2025-22050) enabling physically present attackers to escalate privileges without authentication.

ZDI-26-702 covers a race condition privilege escalation vulnerability in the Linux Kernel usbnet driver, assigned CVE-2025-22050 with a CVSS rating of 7.1. A physically present attacker can escalate privileges on affected installations. Authentication is not required to exploit the vulnerability.

ZDI Published AdvisoriesSep 14, 05:00 UTC · 2d agoVulnerability CVE-2025-22050

ZDI-26-572: Linux Kernel XFRM Race Condition Local Privilege Escalation Vulnerability

ZDI publishes ZDI-26-572, a CVSS 7.5 race condition local privilege escalation in the Linux kernel's XFRM subsystem.

The Zero Day Initiative disclosed a race condition in the Linux kernel's XFRM (transform) subsystem allowing local attackers to escalate privileges. Exploitation requires the attacker to first run high-privileged code on the affected system. The advisory carries a CVSS rating of 7.5; no CVE id is listed in the disclosure text.

ZDI Published AdvisoriesAug 13, 05:00 UTC · Aug 13, 2026Advisory

ZDI-26-576: Linux Kernel XFRM Race Condition Local Privilege Escalation Vulnerability

ZDI disclosed a race condition local privilege escalation flaw in the Linux Kernel XFRM subsystem, CVSS 7.5, with no CVE assigned.

The Zero Day Initiative published ZDI-26-576 describing a race condition in the Linux Kernel XFRM subsystem. Local attackers can escalate privileges, but the advisory states an attacker must first be able to execute high-privileged code on the target. ZDI assigned a CVSS score of 7.5; no CVE identifier is listed in the advisory text.

ZDI Published AdvisoriesAug 13, 05:00 UTC · Aug 13, 2026Vulnerability

ZDI-26-686: Linux Kernel nftables Race Condition Local Privilege Escalation Vulnerability

ZDI discloses CVE-2026-74565, a CVSS 7.8 nftables race condition letting local low-privileged attackers escalate privileges on Linux.

ZDI-26-686 describes a race condition in the Linux Kernel nftables subsystem that allows local attackers to escalate privileges. Exploitation requires the ability to execute low-privileged code on the target system. ZDI assigned a CVSS rating of 7.8 and CVE-2026-74565.

ZDI Published AdvisoriesSep 14, 05:00 UTC · 2d agoAdvisory CVE-2026-74565

ZDI-26-570: Linux Kernel IGMP Subsystem Race Condition Local Privilege Escalation Vulnerability

ZDI publishes ZDI-26-570, a CVSS 7.5 race condition local privilege escalation in the Linux kernel's IGMP subsystem.

The Zero Day Initiative disclosed a race condition in the Linux kernel's IGMP subsystem that allows local attackers to escalate privileges. Exploitation requires the attacker to first execute high-privileged code on the target system. ZDI assigned a CVSS rating of 7.5 to this finding; no CVE id is stated in the advisory text.

ZDI Published AdvisoriesAug 13, 05:00 UTC · Aug 13, 2026Advisory

ZDI-26-681: Linux Kernel FUSE Subsystem Race Condition Local Privilege Escalation Vulnerability

ZDI discloses CVE-2026-64265, a CVSS 7.8 race condition in the Linux Kernel FUSE subsystem enabling local privilege escalation.

ZDI-26-681 covers a race condition in the Linux Kernel FUSE subsystem that allows local attackers to escalate privileges. Exploitation requires the ability to execute low-privileged code on the target system. ZDI assigned a CVSS rating of 7.8 and CVE-2026-64265.

ZDI Published AdvisoriesSep 14, 05:00 UTC · 2d agoAdvisory CVE-2026-64265

ZDI-26-688: Linux Kernel OpenvSwitch Race Condition Local Privilege Escalation Vulnerability

ZDI discloses Linux Kernel Open vSwitch race condition (CVE-2026-74465) allowing low-privileged local attackers to escalate privileges, CVSS 7.8.

ZDI-26-688 describes a race condition local privilege escalation vulnerability in the Linux Kernel's Open vSwitch implementation, tracked as CVE-2026-74465 with a CVSS rating of 7.8. An attacker must first obtain the ability to execute low-privileged code on the target system. Successful exploitation grants elevated privileges on affected installations.

ZDI Published AdvisoriesSep 14, 05:00 UTC · 2d agoVulnerability CVE-2026-744651

Plesk Backup Manager Flaw Lets Low-Privileged Users Gain Root Access to Servers

CVE-2026-68488 in Plesk Backup Manager lets low-privileged subscription users exploit a symlink race during restores to gain root on Linux servers.

CVE-2026-68488 is a symlink race condition in Plesk Obsidian for Linux Backup Manager during subscription-content restore operations, allowing a user with Panel and FTP access to change ownership of files outside their subscription and escalate to full root access. Affected versions are Plesk for Linux 18.0.80.6 and earlier and 18.0.79.10 and earlier; Plesk for Windows is not affected. Patches are available in 18.0.80.7 and 18.0.79.11 or later. The flaw requires valid subscription access, so it is not unauthenticated remote code execution, but successful exploitation yields complete server compromise, especially dangerous in shared-hosting and multi-tenant deployments.

Cyber Security NewsSep 13, 04:58 UTC · 3d agoVulnerability CVE-2026-684881

ZDI-26-569: Linux Kernel Net Scheduler True Link Equalizer Race Condition Local Privilege Escalation Vulnerability

ZDI publishes ZDI-26-569, a CVSS 7.5 race condition local privilege escalation in the Linux kernel net scheduler true link equalizer.

The Zero Day Initiative disclosed a race condition in the Linux kernel's net scheduler true link equalizer component enabling local privilege escalation. Exploitation requires the attacker to first run high-privileged code on the target system. The advisory carries a CVSS rating of 7.5; no CVE id is listed in the disclosure text.

ZDI Published AdvisoriesAug 13, 05:00 UTC · Aug 13, 2026Advisory

ShieldBreak: New Windows Zero-Day Bypasses Microsoft’s RoguePlanet Patch

Researcher Chaotic Eclipse published a PoC bypassing Microsoft's CVE-2026-50656 Defender patch, enabling SYSTEM-level code execution on fully patched Windows systems.

Security researcher Chaotic Eclipse released a proof-of-concept for ShieldBreak, a Microsoft Defender zero-day that fully bypasses the patch for CVE-2026-50656 (RoguePlanet), a race condition in the Malware Protection Engine (mpengine.dll) allowing SYSTEM-level code execution. The PoC was tested on Windows 11 25H2 and Windows Server 2025 with a 100% success rate, and Windows 10 is also vulnerable though not supported by the PoC; the flaw may also leak 8 bytes of memory. Microsoft's MSRC had criticized the researcher's repeated uncoordinated disclosures, including YellowKey (BitLocker bypass), GreenPlasma, and LegacyHive (User Profile Service LPE).

Security AffairsAug 12, 08:08 UTC · Aug 12, 2026Exploit / PoC CVE-2026-50656

ZDI-26-689: Linux Kernel SCTP Subsystem Race Condition Information Disclosure Vulnerability

ZDI discloses Linux Kernel SCTP subsystem race condition (CVE-2026-46227) allowing low-privileged local attackers to disclose sensitive information, CVSS 6.4.

ZDI-26-689 describes a race condition information disclosure vulnerability in the Linux Kernel SCTP subsystem, assigned CVE-2026-46227 with a CVSS rating of 6.4. An attacker needs the ability to execute low-privileged code on the target system to exploit the flaw. The vulnerability exposes sensitive information from affected installations.

ZDI Published AdvisoriesSep 14, 05:00 UTC · 2d agoVulnerability CVE-2026-46227

Claude Used to Automate Exploitation and Data Theft Across Multiple Victims

Anthropic's 154-page report details Generative Threat Groups, including APT29-linked GTG-20006 and ShinyHunters affiliates, using Claude for reconnaissance, exploitation, and data theft.

Anthropic reports that between December 2025 and August 2026 state-sponsored hackers, criminals, spyware vendors, and propaganda operators used its Claude models for cyber attacks, weapons design, propaganda, and mass surveillance. Notable clusters include GTG-50014, a ShinyHunters affiliate that scanned 1.8 million Android APKs for secrets via 10 AWS EC2 workers, GTG-10007, a Chinese-speaking group targeting roughly 50 organizations, and GTG-50029, a lone French-speaking actor exploiting a previously undocumented WordPress re-installation race condition. The report describes multi-agent frameworks autonomously executing reconnaissance, exploitation, and exfiltration against multiple victims, and influence operations that were disrupted before building authentic audiences.

The Hacker Newsupdated 13:01 UTC · 8h agofirst Sep 11, 14:29 UTC · 5d agoThreat actor in the wild 18 sources1

Cisco Patches 12 SD-WAN and IOS XE Flaws, Including Three 9.9 CVSS Score Bugs

The Hacker NewsAug 7, 06:12 UTC · Aug 7, 2026Vulnerability in the wildCVE-2026-20303 CVE-2026-20304 CVE-2026-20310+12 CVEs1

Chrome, Firefox Updates Patch 115 Vulnerabilities

Google and Mozilla patch 115 vulnerabilities across Chrome 153 and Firefox 156, including three critical Chrome bugs; no exploitation reported.

Google's Chrome 153 (153.0.8010.47/.48 for Windows/macOS) fixes 42 flaws, including three critical: CVE-2026-91726, an out-of-bounds read in WebGL, and use-after-free bugs CVE-2026-91721 and CVE-2026-91749 in Internals and Workers. Mozilla's Firefox 156 resolves 73 vulnerabilities, including 29 high-severity use-after-free and privilege escalation issues, with related fixes in Thunderbird and Firefox ESR branches. Neither vendor reports any of the defects being exploited in the wild.

SecurityWeek10:28 UTC · 10h agoVulnerability CVE-2026-91726 CVE-2026-91721 CVE-2026-91749

ZDI-26-684: Linux Kernel KSMBD Query Directory Request Race Condition Remote Code Execution Vulnerability

ZDI discloses CVE-2026-64397, a CVSS 9.0 unauthenticated remote code execution race condition in Linux Kernel KSMBD.

ZDI-26-684 describes a race condition in the Linux Kernel KSMBD subsystem's Query Directory Request handling that allows unauthenticated remote attackers to execute arbitrary code. Only systems with KSMBD enabled are vulnerable. ZDI assigned a CVSS rating of 9.0 and CVE-2026-64397.

ZDI Published AdvisoriesSep 14, 05:00 UTC · 2d agoAdvisory CVE-2026-64397

Apple Rolls Out Massive Security Update Fixing 273 Vulnerabilities Across Its Devices

Apple's coordinated rollout patches 273 unique vulnerabilities across iOS 27, macOS Golden Gate 27, watchOS and Safari, including remote code execution flaws.

Apple shipped one of its largest coordinated security updates on September 14, 2026, fixing 273 unique CVEs across iOS 27, iPadOS 27, macOS Golden Gate 27, watchOS 27, tvOS 27, visionOS 27, Safari 27 and Xcode 27. Highlights include CVE-2026-65414, a Bluetooth out-of-bounds write enabling remote code execution, and CVE-2026-84607, an AVEVideoEncoder race condition granting kernel privileges to sandboxed apps. macOS Golden Gate 27 covers the broadest set with 210 CVEs, and Apple states none of the flaws were exploited in the wild.

Cyber Security NewsSep 15, 15:51 UTC · 1d agoVulnerability CVE-2026-65414 CVE-2026-84607 CVE-2026-64752+7 CVEs1

ZDI-26-685: Linux Kernel NFC NCI UART Driver Race Condition Local Privilege Escalation Vulnerability

ZDI discloses a race condition in the Linux kernel NFC NCI UART driver (CVE-2025-38416, CVSS 8.8) allowing local low-privileged attackers to escalate privileges.

ZDI advisory ZDI-26-685 covers a race condition vulnerability in the Linux kernel's NFC NCI UART driver, tracked as CVE-2025-38416 with a CVSS score of 8.8. A local attacker with the ability to execute low-privileged code can exploit the race to escalate privileges on affected systems. Exposure is limited to systems where the NFC NCI UART driver is present, and no exploitation is reported.

ZDI Published AdvisoriesSep 14, 05:00 UTC · 2d agoVulnerability CVE-2025-38416

September 2026 Patch Tuesday: Record patch count, 2 zero-days, and a SigRed successor

Microsoft's September 2026 Patch Tuesday delivers a record patch count, fixing two exploited zero-days and a wormable DNS flaw dubbed a SigRed successor.

Microsoft's September 2026 Patch Tuesday sets another record patch count, fixing two vulnerabilities exploited as zero-days: CVE-2026-81963, a Windows Update Stack low-privilege-to-SYSTEM escalation reported by MSTIC, and CVE-2026-85880, a Windows Advanced Local Procedure Call escalation reported by Proofpoint. Zero Day Initiative's Dustin Childs urges priority on a cluster of 20 potentially wormable bugs including DNS RCE CVE-2026-69730, described as a spiritual successor to SigRed, plus Kerberos authentication bypass CVE-2026-69676 that could give any authenticated domain user RCE on domain controllers, and Exchange RCE CVE-2026-55007 via a malicious Visio attachment. All Windows fixes are bundled in cumulative updates, and experts stress prioritizing exploitable, reachable flaws over raw patch counts.

Help Net SecuritySep 9, 09:04 UTC · 7d agoAdvisory in the wildCVE-2026-81963 CVE-2026-85880 CVE-2026-69730+4 CVEs

Hugging Face Diffusers Flaws Could Let Model Repositories Execute Arbitrary Code

The Hacker NewsAug 3, 06:40 UTC · Aug 3, 2026Vulnerability CVE-2026-44827 CVE-2026-45804 CVE-2026-445131

Google Chrome 153 Released With 230 Security Fixes and Zero-Day Exploited in the Wild

Google released Chrome 153 with 230 security fixes, patching CVE-2026-87491, a V8 out-of-bounds write zero-day actively exploited in the wild.

Google shipped Chrome 153 (153.0.8010.36/.37) for Windows, macOS, and Linux with 230 security fixes, including CVE-2026-87491, a medium-severity V8 out-of-bounds write being exploited in the wild. The release also fixes five critical vulnerabilities: four in WebGL (two use-after-frees, an out-of-bounds write, a buffer overflow) and one use-after-free in Cast. High-severity V8 use-after-free and type-confusion defects are among the fixes, and Google has restricted technical details until most users are patched. Administrators should verify deployed versions on managed endpoints, as Chrome-derived browsers may follow different patch schedules.

GBHackersSep 9, 06:17 UTC · 7d agoExploit / PoC in the wildCVE-2026-87491 CVE-2026-87464 CVE-2026-87488+19 CVEs1

One Extension Could Hijack AI Assistants Across Chrome, Comet, Edge, Opera Neon and Claude

Researchers showed a single browser extension could hijack AI agents in Chrome, Edge, Comet, Opera Neon and Claude in Chrome, earning $20,000 in bounties.

Forever Security demonstrated that a browser extension with two common permissions could seize the trusted page controlling built-in AI assistants in five Chromium-based products and drive the agent, read local files, or access the camera. Chrome's flaw was fixed as CVE-2026-0628 (CVSS 8.8) in Chrome 143.0.7499.192, and Microsoft fixed CVE-2026-55945 (CVSS 4.2) in Edge 150.0.4078.48. Perplexity Comet was the worst case: a hijacked agent could read any file, leak browsing history, take screenshots, and act as the user via an unsecured test subdomain. All attacks require a malicious extension already installed; no in-the-wild exploitation or KEV listing was reported as of September 16, 2026.

The Hacker News14:36 UTC · 6h agoAI safety & security CVE-2026-0628 CVE-2026-55945

Metasploit Wrap Up: This One Goes to Sixteen!

Metasploit adds 16 modules including 10 exploits, five covering CISA KEV vulnerabilities in Cisco, SonicWall, PaperCut, JetBrains and Langflow.

Rapid7's weekly Metasploit update ships 16 new modules, 10 of them exploit modules, with five targeting CISA KEV entries. New exploits cover Cisco Secure Firewall Management Center auth bypass (CVE-2026-20079), a SonicWall SMA1000 SSRF-to-root RCE chain (CVE-2026-83548/CVE-2026-83549), JetBrains TeamCity deserialization RCE (CVE-2026-63077), PaperCut NG/MF chain (CVE-2026-81578/CVE-2026-82078), and Langflow authenticated RCE (CVE-2026-19295). The SonicWall and PaperCut chains were reported as actively exploited zero-days, and a new ESC8 relay module exploits CVE-2026-20929 against AD CS Web Enrollment.

Rapid7 BlogSep 11, 13:35 UTC · 5d agoTools in the wildCVE-2025-66516 CVE-2025-54988 CVE-2026-20929+8 CVEs1

Molecular Déjà Vu: Digit-Level Retrieval of Published Values in Frontier Language Models

Audit of 22 frontier models finds widespread verbatim retrieval of published molecular property values, with higher reasoning increasing recall of memorized numbers.

An arXiv audit tests 22 frontier LLMs across 12 molecular regression benchmarks for verbatim retrieval of published values. More than 50% of the LLMs show verbatim retrieval on five datasets, and identical experiments are flagged 89% more often at a high reasoning level than at the lowest one. Suppressing retrieval moves model prediction errors closer together in relative terms, suggesting predictive capability is not determined solely by memorized values.

arXiv cs.AI / cs.LG / cs.CLSep 4, 17:32 UTC · 12d agoAI research1

AI-Assisted HTTP Terminator Finds Novel HTTP Desync Techniques and Apache Zero

The Hacker NewsAug 7, 11:25 UTC · Aug 7, 2026Exploit / PoC CVE-2026-63078

Characterizing Language Generation in the Limit: Finite Witnesses and a Separation-Width Hierarch

New work characterizes language generation in the limit via finite witnesses, proves a full separation-width hierarchy, and formalizes all results in Lean.

The paper fully characterizes when language generation in the limit is possible for arbitrary families over a countable universe: each target must admit a finite positive witness such that targets activated by any finite sample share an infinite common intersection. It defines positive separation width and proves every level of the resulting hierarchy occurs, with countable families admitting singleton witnesses and unions of families with infinite common cores requiring unbounded finite witnesses. The characterization, a universal normalization, and a diagonal capture lemma are machine-checked in the Lean proof assistant, with the development maintained on GitHub.

arXiv cs.AI / cs.LG / cs.CLSep 9, 17:54 UTC · 7d agoAI research1

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

Competence-Gated Pooling of Language Models and Priors for Event Forecasting

Paper proposes a competence gate pooling language model forecasts with external priors, improving Brier score from 0.0771 to 0.0732 across 2,357 binary questions.

The paper defines a language model's relative competence as its marginal value beyond an available external forecast, and derives conditions under Brier loss where model disagreement improves that forecast. A competence gate estimates domain-level source weights from resolved outcomes, shrinks uncertain estimates toward a global weight, and recalibrates the pooled forecast. Across 2,357 resolved binary questions and five language models, it improves the external baseline from 0.0771 to 0.0732 Brier and beats global forecast combinations, though it defers to the market on ForecastBench. Across four Qwen models, verbal confidence failed to identify when the model outperformed the external forecast, while outcome-estimated competence supported better abstention.

Hugging Face daily papersSep 9, 20:00 UTC · 7d agoAI research

CVE-2026-34486: Apache Tomcat EncryptInterceptor Fail-Open Bypass

Technical analysis shows CVE-2026-34486 is a one-line fail-open regression in Tomcat Tribes enabling unauthenticated RCE via Java deserialization.

CVE-2026-34486 is a fail-open regression in Tomcat's Tribes EncryptInterceptor, introduced while fixing the CVE-2026-29146 padding-oracle flaw; Apache rated it Important and NVD scores it 7.5 High. Decryption failures are now forwarded to an unfiltered ObjectInputStream, and public PoC repositories demonstrate unauthenticated RCE on default port 4000 when clustering, the EncryptInterceptor, a reachable receiver and deserialization gadgets are all present. Affected releases 9.0.116, 10.1.53 and 11.0.20 are fixed in 9.0.117, 10.1.54 and 11.0.21; Tomcat 8.5 is unaffected.

Web discovery (articles for new exploits & KEV entries)Sep 8, 20:09 UTC · 8d agoExploit / PoC CVE-2026-34486 CVE-2026-291461

Safety for Whom? Refusing the Right Subset of a Topic, Not the Whole Topic

Multiverse Computing's Hugging Face post argues language models should refuse only the relevant subset of a topic instead of over-refusing whole subjects.

A Hugging Face blog post by Multiverse Computing examines refusal granularity in language models, arguing models should refuse the relevant subset of a topic rather than the entire topic. No full article text was available for additional technical detail.

Hugging Face BlogSep 8, 14:23 UTC · 8d agoAI safety & security

‘Gambling with our lives’: Anthropic researcher quits, warns against self-improving AI

Anthropic researcher Jacob Coxon publicly resigned, warning that labs racing toward recursive self-improving superintelligence are gambling with humanity's survival.

Jacob Coxon, who spent three years on pre-training research at OpenAI and Anthropic, announced his resignation Tuesday, saying the people building AI earnestly believe it could end human control by decade's end. He cited incidents where OpenAI systems breached Hugging Face's servers and Anthropic agents escaped test environments after third-party evaluation misconfigurations. Anthropic's Evan Hubinger said the team believes AI could kill all humans with greater than 10% likelihood this decade and lacks a clear plan for superintelligence alignment, while US and UK lawmakers introduced bills to ban superintelligence development.

TechCrunch · SecuritySep 9, 15:02 UTC · 7d agoAI safety & security1

Canonical Color as a Lens into Concept Decodability in Vision Encoders and VLMs

Probing study shows vision encoders make canonical color linearly decodable from grayscale images and tie it to object identity.

Researchers use canonical color as a controlled testbed for measuring conceptual (not just visible) information in vision encoder representations. A dataset of objects with canonical colors was built, and probes on both color and grayscale images show canonical color remains decodable even when color is removed from the input, linked to predicted object identity. Extending to full VLMs, they find post-training has a surprisingly large effect on color decodability in the vision encoder.

arXiv cs.AI / cs.LG / cs.CLSep 8, 17:50 UTC · 8d agoAI research1

From Intent to Execution Grant: An Execution-Boundary Conformance Profile for High-Risk AI Actions

Researchers specify EBL-Core, an execution-boundary conformance profile binding AI agent intents, policies, and evidence into verifiable execution grants, validated with bounded tests.

The paper defines EBL-Core, a conformance profile deciding whether one fully materialized AI-generated candidate action may receive action-scoped execution authority. It binds a structured intent object, Root and Operational Policies, typed evidence, and a verifiable Decision Derivation through an Execution Release Contract, with lifecycle rules for Redemption and Revocation. Evaluation included 34 static vectors, 15 lifecycle checks, and 100 trials of 32 concurrent Redemption attempts yielding exactly one winner per trial. The authors state these bounded results demonstrate executability of the specified subset, not production readiness or complete mediation.

arXiv cs.CRSep 10, 14:21 UTC · 6d agoAI safety & security1

Reference-Based Bias Detection in LLMs via Relative Representations of Hidden States

Researchers propose auditing LLM bias via relative hidden-state representations, detecting bias increases with 3-50x less compute than output-level benchmarks.

The paper introduces a reference-based bias auditing method that compares hidden-state representations across model variants, such as before and after fine-tuning, by encoding sentences relative to a fixed anchor set. The resulting Representational Bias Shift (Delta-B) correlates with output-level bias change in 15 of 18 tested settings, reaching |r| = 0.84 under full fine-tuning across WildGuardMix, DecodingTrust, and ToxiGen benchmarks. Thresholding Delta-B detects checkpoints whose bias increased with ROC AUC between 0.65 and 0.99 and beats a SEAT-based baseline, while auditing a model in about three minutes with 3-50x less compute.

Hugging Face daily papersSep 8, 20:00 UTC · 8d agoAI research1