Latest threat reports & articles

arxiv_cs_cr · tlp:amber · 7/2/2026, 4:00:00 AM

(A)I Sees What You Don't: Exploiting New Attack Surfaces in Third-Party Mobile Agents arXiv:2607.00333v1 Announce Type: new Abstract: Third-party mobile agents powered by Vision-Language Models (VLMs) have emerged as a promising paradigm for automating smartphone interactions. These agents act as high-privilege decision-makers, perceiving device states through screenshots and executing actions via VLM reasoning, transforming how an agent app interacts with the environment (i…

arxiv_cs_cr · tlp:amber · 7/2/2026, 4:00:00 AM

Detecting Adversarial Evasion Attacks Against Autoencoder-Based Network Intrusion Detection Systems arXiv:2607.01194v1 Announce Type: new Abstract: Evasion attacks deliberately manipulate input to an ML-based system to produce an incorrect prediction while the manipulated input still appears benign. The PANDA framework has demonstrated that adversarial examples developed for the vision domain can be transferred to the network domain by converting packet sequences into invert…

arxiv_cs_cr · tlp:amber · 7/2/2026, 4:00:00 AM

All-out Attack: Optimal Block Withholding Under Pay-Per-Share Scheme arXiv:2607.01209v1 Announce Type: new Abstract: Classical Block Withholding (BWH) attacks have been extensively studied in block-dependent reward schemes, where pool members are compensated upon a block discovery within the pool. However, most contemporary mining pools operate under share-based scheme wherein participants are paid immediately upon submission of valid shares. In this paper, we analyze BWH un…

arxiv_cs_cr · tlp:amber · 7/2/2026, 4:00:00 AM

ReShift: Aha-Moment-Driven Reasoning-Level Backdoor Attacks on Vision-Language Models arXiv:2607.00361v1 Announce Type: new Abstract: Vision--Language Models (VLMs) are increasingly deployed in safety-critical applications, yet remain vulnerable to backdoor attacks. Existing methods primarily manipulate final outputs, often producing reasoning traces that are inconsistent or easily detectable. In this paper, we propose ReShift, the novel aha-moment-driven reasoning-level bac…

arxiv_cs_cr · tlp:amber · 7/2/2026, 4:00:00 AM

A Penny for Your Prompts: Experiments Detecting and Mitigating LLM Usage by Survey Respondents arXiv:2607.00403v1 Announce Type: cross Abstract: Large language models are increasingly used by participants on crowdsourcing platforms when responding to surveys, potentially undermining the validity of collected data. Our study aims to quantify the prevalence of this behavior and investigate methods to detect and prevent it. In a series of surveys (N = 250), we examined conditio…

arxiv_cs_cr · tlp:amber · 7/2/2026, 4:00:00 AM

HARC: Coupling Harmfulness and Refusal Directions for Robust Safety Alignment arXiv:2607.00572v1 Announce Type: cross Abstract: Understanding how aligned LLMs internally represent safety is critical for diagnosing alignment vulnerabilities, as it explains why jailbreaks succeed and informs the design of robust alignment strategies. Prior work shows that aligned LLMs encode harmfulness and refusal as separable directions in the residual stream at prompt-side token positions. …

arxiv_cs_cr · tlp:amber · 7/2/2026, 4:00:00 AM

High-Performance NTT Accelerators for PQC leveraging Unified Redundant Arithmetic and Fine-Tuned Microarchitecture arXiv:2607.00621v1 Announce Type: cross Abstract: Post-quantum cryptography and privacy-preserving technologies are expected to play a central role in future secure communication systems. Lattice-based PQC schemes such as ML-KEM (CRYSTALS-Kyber) and ML-DSA (CRYSTALS-Dilithium) rely heavily on large-degree polynomial arithmetic, making the Number Theoretic Transf…

arxiv_cs_cr · tlp:amber · 7/2/2026, 4:00:00 AM

Federated Sovereign Transport Protocol (FSTP): Verifiable Coordination Without Disclosure arXiv:2607.00213v1 Announce Type: new Abstract: This paper introduces the Federated Sovereign Transport Protocol (FSTP), a synchronization boundary and transport layer for federated networks in which nodes have heterogeneous privacy requirements. Existing federation protocols leave data confinement to operator policy: they define message formats and delivery semantics but impose no stru…

arxiv_cs_cr · tlp:amber · 7/2/2026, 4:00:00 AM

Toward a Unified Security and Privacy Framework for AI-Native 6G Networks arXiv:2607.01019v1 Announce Type: new Abstract: Sixth Generation (6G) communication networks are expected to evolve into AI-native, highly autonomous ecosystems that integrate communication, computing, sensing, and artificial intelligence. While these capabilities enable unprecedented connectivity and intelligent services, they also create a highly heterogeneous security and privacy landscape that cann…

arxiv_cs_cr · tlp:amber · 7/2/2026, 4:00:00 AM

Beyond the Prompt: Jailbreaking Function-Calling LLMs via Simulated Moderation Traces arXiv:2607.00481v1 Announce Type: new Abstract: Jailbreak attacks remain a critical threat to the safe deployment of large language models (LLMs). While prior work has primarily studied attacks and defenses at the prompt level, we show that this prompt-centric paradigm overlooks a structural vulnerability in stateful, function-calling environments. In such applications, developer-defined sc…

arxiv_cs_cr · tlp:amber · 7/2/2026, 4:00:00 AM

SessionBound: Turning Enterprise Task Approval into Budgeted Database Sessions arXiv:2607.00751v1 Announce Type: cross Abstract: Enterprise AI agents are useful for internal analysis, audit, compliance review, and operational investigation, but they create a difficult authorization problem. A manager or data owner may approve a business task, while the agent later generates open-ended SQL below the application layer. Existing systems help identify agents, delegate authority,…

lwn_kernel · tlp:amber · 7/2/2026, 12:18:56 AM

[$] LWN.net Weekly Edition for July 2, 2026 Inside this week's LWN.net Weekly Edition: Front : Xsnow protestware; Git 2.55; Rhombus; kernel hardening; More LSFMM+BPF coverage; 7.2 merge window; Secure Boot certificate expiration; Ceph and Garage; OSPM 2026. Briefs : Akrites; Mageia 10; Git 2.55.0; Podman 6.0; systemd v261; Creative Commons chat; Quotes; ... Announcements : Newsletters, conferences, security updates, patches, and more.

chainalysis · tlp:amber · 7/1/2026, 8:26:45 PM

OFAC Updates ISIS-Khorasan Sanctions with Over 100 Cryptocurrency Wallets Summary OFAC updated its ISIS Khorasan (ISIS-K) designation to include 134 cryptocurrency wallet addresses (131 on TRON and 3 on… The post OFAC Updates ISIS-Khorasan Sanctions with Over 100 Cryptocurrency Wallets appeared first on Chainalysis . OFAC Sanctions 100+ ISIS-K Crypto Addresses Chainalysis Products Crypto Investigations Investigations Solutions Reactor Investigate and trace funds acros…

chainalysis · tlp:amber · 7/1/2026, 7:30:58 PM

Chainalysis Supports Robinhood Chain with Automatic Token Support Chainalysis is excited to announce support for Robinhood Chain, a permissionless layer 2 purpose-built for on-chain financial services and tokenized… The post Chainalysis Supports Robinhood Chain with Automatic Token Support appeared first on Chainalysis . Chainalysis Supports Robinhood Chain with Automatic Token Support Chainalysis Products Crypto Investigations Investigations Solutions Reactor Investi…

lwn_kernel · tlp:amber · 7/1/2026, 5:07:42 PM

[$] Efficient access to local storage for BPF programs When a BPF program is used to filter or redirect packets in the networking subsystem, the program will often want to associate data with each packet as it moves through the kernel. The kernel's local BPF storage API , which associates extra data with some kernel objects, provides a way to do that. (See also the BPF map types that end in STORAGE .) Amery Hung and Jakub Sitnicki led two sessions at the 2026 Linux Storage, …

ars_security · tlp:amber · 7/1/2026, 4:11:39 PM

NASA inspector general suggests Boeing's Starliner will now be a decade late Starliner's certification may be delayed to 2027, 10 years later than Boeing's original schedule. NASA's inspector general released an audit Tuesday of the agency's Commercial Crew Program, and it looks increasingly likely that Boeing's Starliner crew capsule won't be certified for operational flights to the International Space Station until next year. That's just three years before NASA's official…

microsoft_mstic · tlp:amber · 7/1/2026, 4:00:00 PM

Microsoft named a leader in the Frost Radar for cloud and application runtime security Frost & Sullivan names Microsoft a leader as cloud and application security converge into unified, runtime risk reduction. The post Microsoft named a leader in the Frost Radar for cloud and application runtime security appeared first on Microsoft Security Blog . Cloud security is shifting from visibility to contextual risk reduction, extending into the applications, APIs, and workloa…

lwn_kernel · tlp:amber · 7/1/2026, 1:18:52 PM

[$] Secure Boot certificate expiration is here Linux users who have Secure Boot enabled on their systems rely on certificates issued by Microsoft to verify the software used to boot a system is trusted by the user. One of those certificates expired recently, but that will not cause systems that are able to boot to stop doing so. There are situations where the expiration may cause problems, however, and the window for relying on existing signed binaries is shorter than it mig…

lwn_kernel · tlp:amber · 7/1/2026, 1:15:44 PM

Security updates for Wednesday Security updates have been issued by AlmaLinux (coreutils, galera and mariadb11.8, giflib, git-lfs, glibc, httpd, kernel, mariadb10.11, mod_md, perl-Archive-Tar, perl-IO-Compress, perl:5.32, rrdtool, ruby, ruby4.0, and thunderbird), Debian (debian-security-support, librabbitmq, and nginx), Fedora (chromium, collectd, maradns, python-django-haystack, python-jupytext, varnish, varnish-modules, and vmod-querystring), Oracle (firefox, git-lfs, kern…

cisa_alerts · tlp:amber · 7/1/2026, 12:00:00 PM

CISA Adds One Known Exploited Vulnerability to Catalog CISA has added one new vulnerability to its Known Exploited Vulnerabilities (KEV) Catalog , based on evidence of active exploitation. CVE-2026-45659 Microsoft SharePoint Server Deserialization of Untrusted Data Vulnerability This type of vulnerability is a frequent attack vector for malicious cyber actors and poses significant risks to the federal enterprise. Binding Operational Directive (BOD) 26-04: Prioritizing Securi…

checkpoint_research · tlp:amber · 7/1/2026, 10:05:35 AM

Browser-Only Ransomware: From LLM Hallucinations to a Practical Attack Technique Research by: Alexey Bukhteyev Key Takeaways Introduction Over the past several years, large language models have reshaped software development, and malware development has followed the same path. Check Point Research has documented this trend from early experiments showing that AI systems could generate offensive components, to cases of cybercriminals using ChatGPT to create malicious tools…

talos · tlp:amber · 7/1/2026, 10:00:57 AM

Martin Lee: Running through the Arctic (and the threat landscape) Ever wonder how someone goes from studying human viruses to leading cybersecurity teams? In this Humans of Talos, we’re joined by Martin Lee, EMEA Lead, to talk about his journey into the industry. Ever wonder how someone goes from studying human viruses to leading cybersecurity teams? In this Humans of Talos, we’re joined by Martin Lee, EMEA Lead, to talk about his journey into the industry. Martin ta…

talos · tlp:amber · 7/1/2026, 10:00:38 AM

ARToken: Inside an EvilTokens affiliate panel targeting Microsoft 365 Talos has identified "ARToken," a phishing-as-a-service platform that targets Microsoft 365. The ARToken panel exposes 80+ API endpoints for device code phishing, Primary Refresh Token persistence, email access, BEC operations, and SharePoint exfiltration. Cisco Talos identified a fully-featured phishing-as-a-service (PhaaS) operator panel, branded "ARToken," that shares infrastructure, API cont…

unit42 · tlp:amber · 7/1/2026, 1:00:11 AM

Phantom Squatting: AI-Hallucinated Domains as a Software Supply Chain Vector Attackers can exploit LLM domain hallucinations through phantom squatting to target supply chains. Read the analysis to learn more. The post Phantom Squatting: AI-Hallucinated Domains as a Software Supply Chain Vector appeared first on Unit 42 . Phantom Squatting: AI-Hallucinated Domains as a Software Supply Chain Vector Menu Tools ATOMs Security Consulting About Us Under Attack? Threat Research Ce…

huggingface_blog · tlp:amber · 7/1/2026, 12:00:00 AM

Hugging Face and Cerebras bring Gemma 4 to real-time voice AI Hugging Face and Cerebras bring Gemma 4 to real-time voice AI Hugging Face Models Datasets Spaces Buckets new Docs Enterprise Pricing Website Tasks HuggingChat Collections Languages Organizations Community Blog Posts Daily Papers Learn Discord Forum GitHub Solutions Team & Enterprise Hugging Face PRO Enterprise Support Inference Providers Inference Endpoints Storage Buckets Log In Sign Up Back to Articles a]:h…

ars_security · tlp:amber · 6/30/2026, 8:03:14 PM

New attack provides one more reason why AI browsers are a bad idea Telling an LLM that 2 + 2 = 5 is enough to make it follow forbidden instructions. Makers of AI browsers make lofty promises. With a single prompt, users can ask one to find a restaurant in a particular part of town, reserve a table, invite a colleague to lunch, and email a confirmation. These makers are much more reticent about the risks of blurring the once fine line between browsing sites and asking a larg…

microsoft_mstic · tlp:amber · 6/30/2026, 7:00:00 PM

Accelerating the quantum-safe timeline We’re accelerating quantum-safe readiness—and sharing what organizations can do now to transition earlier and with confidence. The post Accelerating the quantum-safe timeline appeared first on Microsoft Security Blog . The quantum-safe timeline has changed For years, planning for post-quantum cryptography (PQC) was framed as a future problem: important, inevitable, but distant. That perspective is evolving as technology advances and or…

huggingface_blog · tlp:amber · 6/30/2026, 6:32:50 PM

ScarfBench: Benchmarking AI Agents for Enterprise Java Framework Migration ScarfBench: Benchmarking AI Agents for Enterprise Java Framework Migration Hugging Face Models Datasets Spaces Buckets new Docs Enterprise Pricing Website Tasks HuggingChat Collections Languages Organizations Community Blog Posts Daily Papers Learn Discord Forum GitHub Solutions Team & Enterprise Hugging Face PRO Enterprise Support Inference Providers Inference Endpoints Storage Buckets Log In Sig…

lwn_kernel · tlp:amber · 6/30/2026, 5:53:49 PM

Creative Commons founders' fireside chat (Creative Commons blog) Dee Harris has published a summary of the recent "fireside chat" featuring Creative Commons founders Hal Abelson, Lawrence (Larry) Lessig, Molly Van Houweling, and Glenn Otis Brown. The chat was to mark the 25th anniversary of Creative Commons and included a look back at its history as well as a look at the landscape today: Twenty-five years ago, a small group of people made a bet. They believed that if you gav…

microsoft_mstic · tlp:amber · 6/30/2026, 4:00:00 PM

​​What’s new in Microsoft Security: June 2026 This month’s updates help security and IT teams strengthen identity and multicloud foundations, protect data wherever it lives, and secure the developer workflows powering AI innovation. The post ​​What’s new in Microsoft Security: June 2026 appeared first on Microsoft Security Blog . As organizations scale AI and agents across environments, security teams need protection that covers every surface. The Microsoft vision is simple…

microsoft_mstic · tlp:amber · 6/30/2026, 3:57:11 PM

Securing AI agents: When AI tools move from reading to acting MCP tool poisoning turns trusted AI agents into a control plane for data loss. Learn how threat actors manipulate tool descriptions to trigger unauthorized actions, and how to detect, contain, and prevent it. The post Securing AI agents: When AI tools move from reading to acting appeared first on Microsoft Security Blog . In this article From reading to acting Attack pattern: MCP tool poisoning in a finance workf…

eset · tlp:amber · 6/30/2026, 2:39:37 PM

This month in security with Tony Anscombe – June 2026 edition Three-day patching deadlines, exposed fuel-tank systems, scams costing billions of dollars, and social media bans for children all gave Tony plenty to unpack in June 2026 This month in security with Tony Anscombe – June 2026 edition Award-winning news, views, and insight from the ESET security community English Español Deutsch Português Français TIPS & ADVICE BUSINESS SECURITY ESET RESEARCH About ESET Researc…

huggingface_blog · tlp:amber · 6/30/2026, 2:39:11 PM

Why Specialization Is Inevitable Why Specialization Is Inevitable Hugging Face Models Datasets Spaces Buckets new Docs Enterprise Pricing Website Tasks HuggingChat Collections Languages Organizations Community Blog Posts Daily Papers Learn Discord Forum GitHub Solutions Team & Enterprise Hugging Face PRO Enterprise Support Inference Providers Inference Endpoints Storage Buckets Log In Sign Up Back to Articles a]:hidden"> Why Specialization Is Inevitable Team Article Publ…

lwn_kernel · tlp:amber · 6/30/2026, 1:09:28 PM

[$] Flexible metaprogramming with Rhombus Lisp-like languages have historically led the world in metaprogramming and flexibility. While many modern languages have adopted the idea of macros, Lisp-like languages such as Racket have continued pushing the envelope, attempting to make macros as easy as possible to incorporate into everyday programs. On the other hand, Lisp's minimal, parenthesis-based syntax can be hard to adapt to — to the point that Lisp is sometimes said to s…

lwn_kernel · tlp:amber · 6/30/2026, 1:03:36 PM

Security updates for Tuesday Security updates have been issued by AlmaLinux (git-lfs, perl-Archive-Tar, perl-IO-Compress, python3.12-urllib3, and runc), Debian (sogo), Fedora (perl-DBI and perl-Socket), Oracle (firefox, freerdp, git-lfs, libsoup, libxml2, mod_md, mysql, perl-Archive-Tar, perl-IO-Compress, python, python3.12-urllib3, rsync, thunderbird, tomcat, xorg-x11-server, and xorg-x11-server-Xwayland), SUSE (389-ds, 7zip, alsa, amazon-ecs-init, amazon-ssm-agent, ansible…

cisa_alerts · tlp:amber · 6/30/2026, 12:00:00 PM

StoneFly Storage Concentrator View CSAF Summary Successful exploitation of these vulnerabilities could allow attackers to gain broad unauthorized access, execute arbitrary commands with root privileges, steal sensitive data, and perform actions on behalf of legitimate users across interconnected systems. The following versions of StoneFly Storage Concentrator are affected: Storage Concentrator <8.0.4.22 (CVE-2026-56415, CVE-2026-55721, CVE-2026-50040) Storage Concentrator…

cisa_alerts · tlp:amber · 6/30/2026, 12:00:00 PM

OFFIS DCMTK Toolkit View CSAF Summary Successful exploitation of these vulnerabilities could allow an attacker to write files, access unauthorized information, exhaust memory, or crash affected DCMTK client or server processes. The following versions of OFFIS DCMTK Toolkit are affected: DCMTK <=3.7.0 (CVE-2026-50003, CVE-2026-50254, CVE-2026-35505, CVE-2026-52868, CVE-2026-44628) CVSS Vendor Equipment Vulnerabilities v3 9.8 OFFIS OFFIS DCMTK Toolkit Improper Limitation of…

cisa_alerts · tlp:amber · 6/30/2026, 12:00:00 PM

Mitsubishi Electric MELSOFT Update Manager SW1DND-UDM-M View CSAF Summary Successful exploitation of these vulnerabilities could allow a local attacker to tamper with or destroy information in the affected product, cause a denial-of-service condition in the affected product, or execute arbitrary code when a specially crafted archive file is decompressed by the 7-Zip component included in MELSOFT Update Manager. The following versions of Mitsubishi Electric MELSOFT Update Man…