Google Cloud and Accenture formed a joint unit, training up to 1,000 forward-deployed engineers to drive enterprise adoption of Gemini Enterprise.
Google Cloud and Accenture launched the Accenture Gemini Enterprise Business Group, a joint unit of forward-deployed engineers helping enterprises adopt Google's AI tools. Google will train up to 1,000 Accenture FDEs to build custom applications on the Gemini Enterprise platform. Ramp data cited by TechCrunch puts Google at roughly 6% of US enterprise AI spending versus Anthropic's 43.5% and OpenAI's 39.7%. Alphabet held $811 billion in purchase commitments as of June 30, against Google Cloud's $24.8 billion second-quarter revenue.
CrowdStrike links previously unseen Slim Spider group to crypto custody secret theft at Brazilian financial institutions since March 2026.
CrowdStrike is tracking Slim Spider, a previously undocumented financially motivated group attacking Brazilian financial institutions since at least March 2026. The actor used custom Bash scripts to steal temporary cloud credentials, exfiltrated digital asset custody secrets, and used Foundry's cast tool to derive Ethereum wallet addresses, while deploying Go-based backdoor MikeDor and an implant impersonating Brazil's SPI instant payment infrastructure. Slim Spider also pivoted to Azure DevOps and Kubernetes clusters and used panels including NEXUS // Scanner, Painel de Emails Entra ID, and Painel Pix for reconnaissance and unauthorized Pix transfers. Separately, Google Threat Intelligence Group and Mandiant disclosed Breeze Comet (CL-CRI-1163), a Portuguese-speaking group since 2024 that breaches Brazilian payment infrastructure to run fraudulent Pix, Boleto, and STR transactions.
Check Point found a cross-account covert channel in ChatGPT sandboxes via shared JFrog Artifactory metadata, enabling session hijacking and Gmail data theft. Now fixed.
Check Point discovered that ChatGPT code-execution containers across different accounts could all reach the same internal JFrog Artifactory instance, whose Item Properties API was readable and writable by all accounts, creating a covert cross-account communication channel. Attackers could plant hidden instructions via pasted prompts, shared chat links, or custom GPTs, then trigger tasks in a victim's session to exfiltrate connected-app data such as Gmail, using ChatGPT's default 'Important actions' setting that permits reads without confirmation. OpenAI confirmed and decommissioned the shared Artifactory instance, closing the channel before publication.
Google switches Chrome to two-week release cycles starting with Chrome 153, shrinking the N-day patch gap as AI-driven threats accelerate.
Chrome officially moved from a four-week to a two-week release schedule with Tuesday's launch of Chrome 153 on desktop, iOS, and Android. Google ties the change to its evolving security strategy, saying faster releases shrink the N-day patch gap as AI tools and community bug reports increase patch volume. Mozilla, Microsoft, and Brave have already adopted the two-week cadence, and Google is also racing AI-native browsers like Brave, Dia, Opera Neon, and Perplexity's Comet while experimenting with AI features in Chrome.
Interconnects surveys new open models—Motif-3, GLM-5.3, Hy4-preview—while analyzing a licensing split: Western labs opening up, Chinese frontier labs getting restrictive.
The roundup covers Motif-3 (MIT license, strong scores for its size), GLM-5.3 (switched from MIT to a custom license with a $10 billion revenue threshold and undefined 'affiliates' clause requiring Z.AI security review), and Tencent's Hy4-preview (competent but prone to overthinking). It also notes dots3-note-prev from RedNote/Xiaohongshu (won IMO 2026 with a perfect score), Qwen3.8-Flash-Next (125B-A6B with GDN and Qwen Sparse Attention), NVIDIA Nemotron-3.5-Lightning-30B-A3B-BF16, and Ling-3.0-flash. The core theme: Google and Meta adopted Apache 2.0 while Chinese frontier labs (Zhipu, Kimi K3, MiniMax M3) adopted restrictive commercial licenses.
Google DeepMind launches AlphaGenome Atlas, a catalog of predicted molecular effects for roughly nine billion single-letter DNA changes across the human genome.
Google DeepMind unveiled AlphaGenome Atlas, a roughly 1-petabyte dataset predicting how each of about nine billion possible single-nucleotide variants affects molecular biology, including non-coding regions that regulate gene behavior. It builds on the AlphaGenome model released in 2025 and adds a Variant Impact Score (AVI) to help researchers rank variants. The catalog is available for noncommercial research via a web portal, the Antigravity platform, and the AlphaGenome interface, with commercial access on Google Cloud planned.
Google's GTIG reports threat actors using autonomous AI agents, credential stealers, and LLMs to steal AI models, API credentials, and harvest thousands of credentials.
Google Threat Intelligence Group says attackers are targeting proprietary AI models across healthcare, government, and media, exfiltrating API credentials, and co-opting victim cloud environments to run unauthorized AI workloads. TeamPCP (Altered Spider/UNC6780) is conducting supply chain compromises of PyPI, npm, and Docker Hub, deploying the SANDCLOCK and DUSTMAKER credential stealers, with DUSTMAKER using AI workspace poisoning and prompt injection for defense evasion. One financially motivated actor used an autonomous multi-agent framework to compromise thousands of third-party credentials in under six hours without human intervention. China-nexus groups UNC6508 and Basin Castle (Mustang Panda) used local open-weight LLMs and commercial LLMs like Gemini, Claude, and Codex for espionage tasks and evading provider monitoring.
Adobe adds in-timeline generative media to Premiere, letting editors generate video and audio clips using Firefly, Veo, Runway, Luma, and Kling models.
Adobe's new Generative Media tool lets Premiere editors highlight empty gaps in the timeline and generate context-aware, editable video, sound effects, music, and soundscapes without leaving the project. Editors can choose among underlying models including Adobe Firefly, Google Veo, Runway, Luma, and Kling. Beta AI audio tools can separate overlapping speakers and duck music under speech, and an AI Assistant is coming to After Effects for plain-language project commands.
Malone Lam's plea hearing approaches in the $240 million bitcoin social engineering theft; the case highlights surging crypto fraud and limited enforcement.
Malone Lam, accused of organizing a social engineering attack that stole over $240 million in bitcoin (4,100+ BTC) from a Washington, D.C. resident in August 2024, has a plea agreement hearing set. Callers impersonating Google and Gemini staff tricked the victim into revealing security codes. Lam and 17 co-defendants spent lavishly before FBI arrests; crypto investment fraud complaints to the FBI rose nearly 50% in 2025 while DOJ disbanded its crypto crimes unit.
Meta's AI ad-detection failed to catch 350+ CSAM video ads on Facebook, Instagram, and Threads, some depicting images of real children.
The Tech Transparency Project found over 250 additional ads containing child sexual abuse material on Meta platforms since August, on top of ~53 previously removed, exceeding 350 total since late last year. Some ads used images of real children, including a European royal family minor and teen influencers, morphed into graphic sexual videos via AI face-swapping. Ads linked to nudification apps from Chinese developers and reached over 29,000 EU accounts plus thousands in the US, UK, Australia, and India.
Google's Mandiant warns extortion crews now steal proprietary AI models, prompts, and research for ransom, while threat actors automate attacks with agentic AI.
Google's AI Threat Tracker details intrusions in which extortion crews exfiltrated healthcare drug research and a proprietary AI model, and stole an AI media company's source code, prompts, skills, model scripts, and secrets, threatening leaks unless paid. TeamPCP (UNC6780), behind large open-source supply chain attacks on PyPI, npm, and Docker Hub since March, used a malicious GitHub Actions workflow to exfiltrate a company's proprietary AI repository. Mandiant also observed a fully autonomous multi-agent credential-harvesting cloud intrusion completed in under six hours, and a China-linked espionage group using Gemini to build a dynamic automated penetration-testing framework.
AP report on scammers' spending spree after a $240 million bitcoin social engineering theft, as ringleader Malone Lam nears a plea agreement.
DataBreaches.net syndicates AP reporting on the $240 million bitcoin theft from a Washington, D.C. resident via social engineering calls impersonating Google and Gemini. Alleged ringleader Malone Lam, 22, faces a plea hearing; the scam network spent stolen funds on cars, jets, and mansions before FBI arrests. Crypto investment fraud complaints to the FBI rose nearly 50% in 2025.
Ex-Google DeepMind researcher Danijar Hafner founded a stealth robotics startup applying world models and model-based reinforcement learning to humanoid agents.
Danijar Hafner, 31, left Google DeepMind in fall 2025 to found a stealth San Francisco startup developing humanoid robots that plan ahead using world models trained via model-based reinforcement learning. His prior work includes PlaNet, Dreamer 2 (first human-level Atari agent in a world model), Dreamer 3 (solved the Minecraft Diamond challenge), Dreamer 4 (learned diamond mining from offline video), and DayDreamer, which let robots adapt to novel situations without task-specific training. The profile covers his career from Google Brain intern to founder aiming to handle unfamiliar real-world environments.
Researchers demonstrate a decryption jailbreak that extracts encrypted reasoning traces from Anthropic, OpenAI, and Google LLM APIs via weaker sibling models.
The paper exploits the fact that encrypted chain-of-thought blocks returned by LLM providers are interchangeable across sessions, users, and models within a provider's ecosystem. Injecting an encrypted trace into a weaker, less-safeguarded model from the same provider forces it to output the trace in plaintext, bypassing anti-distillation mechanisms. Decoding 315,320 reasoning blocks scraped from public repositories recovered 367 PII artifacts and 182 credentials, showing large-scale private data leakage. The flaw also enables hidden hazardous information disclosure and invisible prompt injections embedded in encrypted blocks; mitigations were proposed after responsible disclosure.
Talos tracks a crypto-theft campaign abusing the Google Visualization API for C2, using browser-based ClickFix lures to inject web-skimmer JavaScript.
Criminal actors convinced targets to paste JavaScript into Chrome's address bar or install it in the Tampermonkey extension, injecting a web skimmer into sessions on two cryptocurrency trading websites; Tampermonkey also provides persistence. Since March 2026 the campaign retrieved obfuscated scripts via the Google Visualization API from public Google Sheets documents, hooking the fetch API and replacing cryptocurrency deposit addresses in responses and the clipboard. Lures pose as leaked reports of a nonexistent API flaw at cryptocurrency swap services and spread via Telegram, DarkForums, and paste sites since early October 2025. Talos warns the legitimate-service-abuse techniques could enable wider supply-chain attacks on e-commerce and customer-facing systems.
Researchers show IICL structural jailbreaks generalize to Google Gemini, lifting attack success to 80-100% on harm and financial benchmarks; non-English prompts attenuate it.
The study red-teams two Google Gemini models with Involuntary In-Context Learning (IICL), a structural jailbreak reframing harmful requests as the final cell of a data-labeling task. IICL lifts attack success from at most 6.7% to 80-90% on HarmBench and 97-100% on financial abuse (FinProof), an order of magnitude above prior results on OpenAI's GPT-5.4. Against a compounding hypothesis, forcing IICL output into Spanish, Hindi, or Arabic attenuates the attack in 11 of 12 conditions, attributed to a 'relevance curse' producing lower-quality harmful content in lower-resource languages. Findings replicate under an independent non-Google judge (Cohen's kappa 0.86 over 377 paired verdicts).
Google patches actively exploited Chrome zero-day CVE-2026-85046, a V8 type confusion bug enabling sandboxed code execution via crafted HTML pages.
Google patched 12 Chrome vulnerabilities, including CVE-2026-85046, a CVSS 8.8 type confusion flaw in the V8 JavaScript and WebAssembly engine exploited in the wild. The bug, reported by researcher Salvatore Gulizia (Serotav) on August 4, 2026, allows arbitrary code execution inside the sandbox via a crafted HTML page; the researcher described it as a V8 compiler bug enabling arbitrary read/write on the JavaScript heap. Fixes shipped in Chrome 152.0.7977.82/.83 for Windows and macOS and 152.0.7977.82 for Linux, and this is the sixth Chrome zero-day Google has fixed in 2026.
Fulton County, Georgia — home to Atlanta and over 1 million residents — confirms a cyberattack that disrupted county phone, court, and tax systems.
Fulton County government disclosed a widespread IT outage caused by a cybersecurity incident discovered over the weekend, affecting phone systems, the court system, and the tax system, including firearm and marriage license transactions. County Chairman Robb Pitts said the incident is under active investigation by law enforcement, and the FBI confirmed its involvement. The Tax Commissioner office in downtown Atlanta was closed and library public computers were offline. No group has claimed responsibility, though the article notes recent ransomware attacks on other Georgia local governments, including Forsyth County (AlphV) and Augusta.
SAEScientist-Bench evaluates whether AI agents can autonomously conduct SAE interpretability research in Gemma-2-9B-IT, finding frontier agents trail expert baselines.
SAEScientist-Bench tests if AI agents can act as scientists using SAE tools for autonomous mechanistic discovery, requiring them to design contrastive probes and navigate a Gemma Scope dictionary of 131K+ features in Gemma-2-9B-IT. Across 10 agent configurations and 20 tasks, frontier agents showed genuine discovery capability but remained well behind expert reference features, lagging most in causal steering. Agents frequently misinterpreted experimental measurements even when designing effective contrasts.
SOCRadar disclosed PEEP, a post-exploitation Chromium extension toolkit that turns Chrome and Edge into host-level backdoors via native messaging and C2 polling.
SOCRadar researchers detailed PEEP, a Chromium-based post-exploitation toolkit masquerading as a 'Smart Bookmarks' extension injected directly into Chrome and Edge profiles by forging Secure Preferences integrity values, bypassing Web Store checks. A native-messaging host binary (nm_host.exe) extends the browser agent to host-level command execution, file management, credential theft, and session hijacking, while the extension polls C2 endpoints at 206.237.30.232 or xfjcc.fun every 30 seconds over plaintext HTTP and exfiltrates browsing history, cookies, and tab metadata. PEEP is derived from the open-source RedExt red teaming framework, adds PowerShell persistence scripts, and includes a Linux-targeting Python script, suggesting cross-platform intent. The activity is unattributed but contains Chinese-language artifacts; a C2 /health endpoint showed 34 agent entries and 10 active sessions.
Similarweb data shows ChatGPT regaining chatbot web traffic share to 55.5% while Gemini slipped to 25.6% and Claude grew to 9.3%.
Similarweb figures show ChatGPT's share of AI chatbot website traffic rising from 52.7% three months ago to 55.5%, though it remains far below its 73.3% share a year ago. Google Gemini declined from 27.8% to 25.6% after a brief comeback, while Anthropic's Claude grew from 1.9% to 9.3% year-over-year. DeepSeek (3.4%), Grok (2.4%), Copilot (1.6%), and Perplexity (0.9%) trail the leaders. The data covers website traffic only and excludes mobile and desktop app usage.
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.
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.
Ars Technica probes diffuse accountability behind TeraWulf's $3.2B Lake Mariner AI data center after a June fire exposed safety and job gaps.
A June fire at the Lake Mariner data center in Somerset, New York exposed missing alarms, a nonfunctioning suppression system, and dry hydrants, highlighting how responsibility is split across TeraWulf (owner-operator), Fluidstack (operator), Google (lease guarantees and equity warrants), and Anthropic (compute customer). The article details local concerns over the gap between promised 165 permanent jobs and a projected 35-40, socialized grid costs, and Governor Hochul's moratorium on hyperscaler development. Anthropic's February 2026 pledge to cover electricity price increases applies to the site but leaves other commitments unverified.
Google patched 12 Chrome flaws including in-the-wild V8 zero-day CVE-2026-85046; CISA added it to the KEV catalog.
Google released Chrome 152.0.7977.82/.83 for Windows and Mac (152.0.7977.82 for Linux) fixing 12 vulnerabilities, including CVE-2026-85046, a type confusion flaw in the V8 JavaScript engine being exploited in the wild. CISA added CVE-2026-85046 to its Known Exploited Vulnerabilities Catalog with a patch deadline of September 18, 2026. This is Chrome's sixth zero-day patched in 2026, following CVE-2026-2441, CVE-2026-3909, CVE-2026-3910, CVE-2026-5281 and CVE-2026-11645. The other 11 fixes include use-after-free, out-of-bounds, race condition and input validation flaws in Skia, WebGL, DevTools, Network, Compositing and other components.
Security Affairs weekly newsletter round 593 linking to dozens of stories across breaches, ransomware, malware, APTs, and AI security news.
This is the weekly Security Affairs newsletter aggregation (round 593) with a list of headlines: Manchester Airports Group breach (8.8M people), Aurora ransomware targeting ESXi, active exploitation of PaperCut and SonicWall SMA 1000 zero-days, Chrome V8 zero-day patching, Pegasus spyware of Serbian activists, and DPRK APT activity against South Korea. It is an index page rather than a standalone story.
Help Net Security week-in-review roundup: Claude infostealer lockouts, record Patch Tuesday, McKesson breach, SonicWall zero-days, PaperCut attacks.
This Help Net Security weekly roundup aggregates prior stories: Anthropic locking out Claude users after infostealer-compromised login sessions, August 2026 Patch Tuesday's 398 resolved CVEs, ShinyHunters' claim of 284 million stolen patient records from McKesson, active exploitation of SonicWall SMA 1000 zero-days (CVE-2026-83548/83549), Sangoma Switchvox SQL injection exploitation (CVE-2026-9586), and attackers planting RATs on compromised PaperCut servers. It also covers FBI OAuth phishing warnings, a Spring Ring vishing campaign abusing Microsoft Teams, and a fake Claude Opus 5 GitHub repo delivering RevStealer infostealer.
Three hikers were rescued from Mount Shasta after following Google Gemini's advice to carry insufficient food and water during a multi-day ordeal.
Three young men began their Mount Shasta ascent at 3 a.m. and summited at 7 p.m., far past the recommended noon turnaround, then tried descending in the dark. The Siskiyou County sheriff's office said Gemini advised far less food and water than required as the planned 8-hour climb became a multiday ordeal. The trio spent the night in Mud Creek Canyon and were rescued the next morning by Forest Service rangers and volunteers.
CISA added actively exploited Chrome V8 type-confusion flaw CVE-2026-85046 (CVSS 8.8) to its KEV catalog; Google patched it in Chrome 152.
CISA added CVE-2026-85046 (CVSS 8.8), a V8 type confusion flaw, to its Known Exploited Vulnerabilities catalog, requiring US federal agencies to patch by September 18, 2026. Google fixed it in Chrome Stable 152.0.7977.82/.83 for Windows and Mac and 152.0.7977.82 for Linux, alongside 11 other vulnerabilities. The bug lets a remote attacker execute arbitrary code inside the browser sandbox via a crafted HTML page and is the sixth actively exploited Chrome zero-day of 2026. Researcher Salvatore Gulizia (Serotav) reported it on August 4, 2026.
Cadence pairs Google's 330M-parameter TimesFM-3 foundation model with adaptive arithmetic coding, gaining 13-28% on 2026 demand series over classical predictors.
Cadence is an error-bounded lossy compressor for numeric time series combining the 330M-parameter Google TimesFM-3 foundation model with an adaptive arithmetic coder, guaranteeing a per-sample error bound. On 49 EIA-930 balancing-authority demand series from 2026 it gains 13.3% over the best of six classical predictors and 28.3% on 50 MTA ridership series, winning all 297 series-tolerance pairs with a 21.4% median gain. The paper also reports negative results, including that foundation models add negligible value for lossless coding and that PyTorch predictions are not bit-identical across batch sizes.
Google patched actively exploited Chrome flaw CVE-2026-85046 in Chrome 152.0.7977.82; CISA added it to the KEV catalog, urging updates.
The Canadian Centre for Cyber Security (advisory AV26-883, Update 1) reports Google fixed CVE-2026-85046 in Chrome 152.0.7977.82 and that an exploit exists in the wild. CISA added the vulnerability to its Known Exploited Vulnerabilities database on September 4, 2026. Users and administrators should update Chrome stable desktop channels promptly.
Google's Gemini Spark agent can now edit images, curate albums, and create calendar events in Google Photos for eligible US subscribers.
Google announced that its personal agent, Gemini Spark, can now manage Google Photos libraries, including editing images, curating albums, creating shared albums, converting concert flyer photos into calendar appointments, and running workflows. Google Photos lead Shimrit Ben-Yair said the capabilities roll out over the next few weeks to eligible Gemini AI Pro and Ultra subscribers in the US, in English, with no international timeline given. The move illustrates the industry's push to demonstrate consumer AI value amid criticism that labs have poorly communicated AI's benefits.
AI firm Micro1 bid $12.5 million for Spirit Airlines' 600 million employee email and chat records, challenging Google's $10 million auction win.
After Spirit Airlines entered bankruptcy, Google won an auction for the airline's data trove with a $10 million bid, beating Mercor's $7.5 million offer, but Micro1 has now counter-offered $12.5 million. The data includes about 600 million email and chat records from 17,000 employees, 17 million OneDrive files, 20.5 million SharePoint items, and over 30 million customer service calls, valuable for AI training. Former employees' unions, including the Association of Flight Attendants-CWA representing 5,500 flight attendants, have taken legal action to block the sale over privacy concerns.
CISA added CVE-2026-85046, a Google Chromium V8 type confusion vulnerability with evidence of active exploitation, to its KEV Catalog.
On September 4, 2026, CISA added one vulnerability to the Known Exploited Vulnerabilities Catalog: CVE-2026-85046, a type confusion flaw in Google Chromium's V8 engine, added based on evidence of active exploitation. CISA notes type confusion is a frequent attack vector and poses significant risk to the federal enterprise. Under Binding Operational Directive 26-04, Federal Civilian Executive Branch agencies must prioritize remediation of KEV vulnerabilities on exposed assets and check for pre-patch compromise; all organizations are urged to prioritize the patch.
Instagram is mislabeling ordinary edited photos as AI Content while some AI imagery goes unlabeled, repeating a 2024 detection failure.
The Verge documents weeks of erroneous AI Content labels on Instagram, including images edited only with Canva's Background Remover or an iPhone Photos app, while some generative images escape tagging. Canva said some of its assistive AI tools were being tagged as generative and claims the issue is fixed, though users still report tagging. Meta scans IPTC and C2PA metadata and uses signals like Google's SynthID, but remains vague about detection criteria; one tester found only Meta AI-created or edited images reliably triggered labels, and an image-poisoned photo was tagged. A similar mislabeling wave hit Instagram in 2024.
DHS subpoenaed REI for all Minneapolis-area customers who bought a specific green beanie since 2024, part of an investigation into 39 ICE protest defendants.
Court filings allege Homeland Security Investigations agents subpoenaed REI in March for transaction records of all persons in the greater Minneapolis–St. Paul area who purchased a specific dark green beanie since 2024. The subpoena was one of 92 sent in a federal case against 39 people, including journalists, who attended an ICE protest at a church. Companies responded differently: T-Mobile handed over six months of a defendant's call and text logs, Google refused a request for YouTube viewers, Reddit withdrew after a First Amendment objection, and Meta pushed back on at least one summons. The 1509 customs summonses require no judicial oversight, and the total number issued under the Trump administration is unknown.
Google patched 12 Chrome flaws including actively exploited V8 type confusion zero-day CVE-2026-85046 (CVSS 8.8), enabling sandboxed remote code execution via crafted webpages.
Google released Chrome Stable 152.0.7977.82/.83 for Windows and Mac and 152.0.7977.82 for Linux, fixing 12 vulnerabilities including CVE-2026-85046, a type confusion in V8 with a CVSS score of 8.8. Google says an exploit exists in the wild, making this the sixth actively exploited Chrome zero-day of 2026. The bug affects the JavaScript and WebAssembly engine and allows arbitrary code execution inside the browser sandbox via a specially crafted HTML page. Researcher Salvatore Gulizia (Serotav) reported the flaw on August 4, 2026 and received a $1,000 bounty.
Google patches actively exploited Chrome V8 zero-day CVE-2026-85046 (CVSS 8.8); CISA adds it to KEV with a September 18 deadline.
Google fixed 12 Chrome vulnerabilities, including CVE-2026-85046, a high-severity (CVSS 8.8) type confusion bug in the V8 JavaScript and WebAssembly engine. The flaw, reported by researcher Salvatore Gulizia on August 4, 2026, allows remote arbitrary code execution inside the browser sandbox via a crafted HTML page. Google confirmed the exploit exists in the wild, and CISA added the CVE to its KEV catalog, requiring Federal Civilian Executive Branch agencies to patch by September 18, 2026. This is the sixth actively exploited Chrome zero-day patched in 2026.
Elastic compares SIEM data egress cost, latency, and fidelity across CrowdStrike, Microsoft, Google, and Splunk, arguing vendors engineer lock-in.
Elastic Security Labs published an opinion piece comparing how major SIEM and security vendors handle data egress, based on each vendor's public documentation as of September 2026. It rates CrowdStrike Falcon Data Replicator and Palo Alto Networks XSIAM Event Forwarding as restricted (paid add-ons with batch delays), Microsoft as partially open, Splunk as open, and Elastic as open with no export license. The piece argues frictionless ingestion paired with licensed or delayed egress is an intentional lock-in business model, and cites CrowdStrike's 2026 Global Threat Report eCrime breakout time of 29 minutes to argue real-time telemetry access is now essential.
Malwarebytes threat data shows scammers tailor platforms per scam, with the web as top channel, Friday midday peaks, and MrBeast the most impersonated person.
Malwarebytes analyzed its threat data from April 15 to July 14, 2026 across more than 20 scam categories, finding scammers match platforms to scam types: job scams via email, romance scams via social media, and tech support scams via phone. The web is the top delivery channel ahead of email and SMS, and Malwarebytes says it blocks about 500,000 phishing sites a day. Scam texts peak at 12:00 pm ET, roughly 874% above the quietest hour, with volume peaking on Fridays about 50% higher than the start of the week. MrBeast (Jimmy Donaldson) appears in about 30% of impersonation scams, and the most impersonated brands are Google, Microsoft, Apple, Roblox and Amazon.
Actively Exploited V8 Type Confusion in Google Chrome (CVE-2026-85046)
Google Chrome versions prior to 152.0.7977.82 contain a type confusion flaw (CWE-843) in the V8 JavaScript engine, which mishandles object types during engine operations (public proof-of-concept writeups indicate it is reachable through array sorting and WebAssembly-related code paths). A remote attacker triggers the flaw simply by getting a user to open a crafted HTML page, with no privileges or authentication required. Successful exploitation lets the attacker execute arbitrary code inside the browser's sandbox, and public reporting shows it being chained with Windows zero-days (the 'BlueMoon' exploit kit) by Chinese espionage groups for broader compromise. Any user of an unpatched Chrome or another build embedding the affected V8 engine is exposed. The vulnerability is a zero-day that was actively exploited in the wild before patching, was added to CISA's KEV on 2026-09-04, and has five public proof-of-concept references.
· Google Chrome prior to 152.0.7977.82 · Google Chromium V8 V8 engine versions bundled with Chrome prior to 152.0.7977.82 KEV PoC ×5mass
Actively Exploited Out-of-Bounds Write in Google Chrome V8
CVE-2026-87491 is an out-of-bounds write (CWE-787) in the V8 JavaScript engine in Google Chrome, fixed in Chrome 153.0.8010.36, which Google shipped alongside roughly 230 other security fixes. An attacker can trigger the flaw remotely by luring a user (user interaction required) into opening a crafted HTML page that corrupts memory in V8. Successful exploitation allows the attacker to execute arbitrary code inside the Chrome browser sandbox, which constrains but does not eliminate the impact. All Google Chrome users running versions prior to 153.0.8010.36 are affected; because the flaw resides in V8, CISA tracks it as 'Google Chromium V8', and other Chromium-based browsers may inherit the fix in their own updates. The flaw is being actively exploited in the wild — it is the seventh actively exploited Chrome zero-day of 2026 and was added to CISA's KEV catalog on 2026-09-09 — though no public proof-of-concept is known and ransomware use is unknown.
· Google Chrome (V8 JavaScript engine; tracked by CISA as 'Google Chromium V8') prior to 153.0.8010.36 KEVmass
Heap-Based Buffer Overflow in Windows ALPC Enables Local Privilege Escalation
CVE-2026-85880 is a heap-based buffer overflow in Windows Advanced Local Procedure Call (ALPC), the Windows mechanism for local inter-process communication. An authorized local attacker can trigger the overflow by submitting crafted input over ALPC, corrupting heap memory in the component that handles the request. Successful exploitation allows the attacker to execute code with elevated privileges, typically gaining SYSTEM-level control of the local host, which is especially valuable as a post-exploitation or sandbox-escape step. Affected products include Windows 10 (1607, 1809, 21H2, 22H2) and Windows Server 2012, 2016, 2019, and 2022, meaning most on-premises Windows estates are in scope. The flaw was fixed in Microsoft's record 974-CVE September 2026 Patch Tuesday and was added to CISA's KEV on 2026-09-08, confirming exploitation in the wild; press reports describe Windows zero-days being chained with a Chrome zero-day in 'BlueMoon' kit attacks, though the data does not explicitly confirm this CVE is the Windows flaw in that chain.
· Microsoft Windows 10 1607, 1809, 21H2, 22H2 · Microsoft Windows Server 2012, 2016, 2019, 2022 KEVmass
Use-After-Free in Google Chromium CSS Rendering Exposes Chrome, Edge, Opera Users
CVE-2026-2441 is a use-after-free (CWE-416) in Google Chromium's CSS handling that a remote attacker can trigger by getting a user's browser to process a crafted HTML page, potentially corrupting the heap. Successful exploitation yields a memory-corruption primitive in the browser; CVSS scoring is not yet available, but Chromium memory-safety flaws of this class can range from crashes to potential code execution depending on how the corruption is leveraged. Anyone running Chromium or a Chromium-based browser — Google Chrome, Microsoft Edge, Opera, and numerous embedded/branded browsers — is potentially affected, making the exposed population effectively all modern browser users. CISA added the flaw to the Known Exploited Vulnerabilities catalog on 2026-02-17, confirming it is being exploited in the wild; EPSS assigns a 22% probability of exploitation within 30 days (98th percentile), no public PoC is known, and any ransomware association is unknown. This lands amid an accelerating series of actively exploited Chrome zero-days in 2026 described in recent reporting, making rapid patching urgent.
· Google Chromium · Google Chrome (Chromium-based) KEV PoC mass
Out-of-Bounds Write in Google Chrome's Skia Rendering Engine
CVE-2026-3909 is an out-of-bounds write (CWE-787) in the Skia graphics rendering library used by Google Chrome, which Google patched in Chrome 146.0.7680.75. A remote attacker can trigger the flaw by luring a user to visit a crafted HTML page, causing out-of-bounds memory access in the browser process; the High CVSS 8.8 score with high confidentiality, integrity, and availability impact indicates memory corruption with potentially serious consequences. All Google Chrome installations running versions prior to 146.0.7680.75 are affected, and because the vulnerability resides in the Skia engine, the vulnerable code is present in every Chrome install. CISA added the flaw to its Known Exploited Vulnerabilities catalog on 2026-03-13, indicating exploitation in the wild, though no public proof-of-concept is known and ransomware use is unknown. EPSS currently assigns a 1.6% probability of exploitation within the next 30 days (75th percentile).
· Google Chrome (Skia rendering engine) all versions prior to 146.0.7680.75 · Google Skia (as shipped in Google Chrome, per CISA) prior to Chrome 146.0.7680.75 KEVmass
Out-of-Bounds Read/Write in Google Chrome V8 Enables In-Sandbox Code Execution
CVE-2026-11645 is an out-of-bounds read and write (CWE-125/CWE-787) in V8, the JavaScript engine used by Google Chrome and Chromium. A remote attacker triggers the flaw by luring a user to a crafted HTML page, where malicious script causes V8 to read and write outside allocated memory buffers. Successful exploitation allows the attacker to execute arbitrary code inside the browser's security sandbox, providing limited privileges within that process rather than full system compromise. All Google Chrome versions prior to 149.0.7827.103 are affected, along with the Chromium V8 component identified by CISA. The flaw is being exploited in the wild: CISA added it to the Known Exploited Vulnerabilities catalog on 2026-06-09, and related headlines describe an actively exploited Chrome V8 zero-day, though no public proof-of-concept is known and ransomware use is unconfirmed.
· Google Chrome prior to 149.0.7827.103 · Google Chromium V8 (as shipped in Chrome) prior to the 149.0.7827.103 Chrome release KEVmass
Out-of-Bounds Memory Flaw in Google Chrome V8 Allows Sandboxed Code Execution
CVE-2026-3910 is an out-of-bounds memory flaw — CISA classifies it as improper restriction of operations within the bounds of a memory buffer — caused by an inappropriate implementation in the V8 JavaScript and WebAssembly engine used in Google Chrome. A remote attacker triggers it by luring a user to a crafted HTML page, so exploitation requires user interaction but no privileges or special conditions. Successful exploitation allows the attacker to execute arbitrary code inside the Chrome sandbox, which confines the compromise to the renderer process rather than the full system, but still exposes data and credentials handled within the browser session. Anyone running Google Chrome prior to 146.0.7680.75 — or a Chromium-based browser that embeds the vulnerable V8 code — is affected. The flaw is being actively exploited in the wild: CISA added it to the Known Exploited Vulnerabilities catalog on 2026-03-13, and it comes amid a string of actively exploited Chrome V8 zero-days that Google has fixed during 2026.
· google chrome prior to 146.0.7680.75 (fixed in 146.0.7680.75) KEVmass
Use-After-Free in Google Chrome's Dawn (WebGPU) Component Enables Arbitrary Code Execution
CVE-2026-5281 is a use-after-free vulnerability in Dawn, the WebGPU implementation in Google Chrome, that Google patched in Chrome 146.0.7680.178. It is triggered when a remote attacker serves a crafted HTML page and can leverage it after having already compromised the Chrome renderer process, escalating to arbitrary code execution beyond the initial foothold. Because the flaw requires a compromised renderer as a starting point, it is typically chained with another bug (such as a renderer-exploiting issue) to break out to arbitrary code execution with real impact on confidentiality, integrity, and availability. Any user or organization running Google Chrome on a version prior to 146.0.7680.178 is affected. The flaw is confirmed as exploited in the wild: CISA added it to the Known Exploited Vulnerabilities catalog on 2026-04-01, and its EPSS score of 4.9% (92nd percentile) indicates a meaningful near-term exploitation probability.
· Google Chrome prior to 146.0.7680.178 · Google Dawn (WebGPU implementation bundled in Chrome) as shipped in Chrome prior to 146.0.7680.178 KEVmass
Local Privilege Escalation via Link Following in Windows Update Stack
CVE-2026-81963 is a link-following flaw (CWE-59, improper link resolution before file access) in the Microsoft Windows Update Stack, in which the component fails to correctly resolve file links before opening them. A local attacker with low privileges can plant or manipulate a link (symlink/junction) that the privileged update stack follows during operation, redirecting its file access to an attacker-controlled target. The result is local privilege escalation — CVSS 3.1 rates this 7.8 (high) with high confidentiality, integrity, and availability impact — allowing an authorized local user or malware already on the machine to gain elevated rights. Affected products are Windows 11 23H2, 24H2, 25H2, and 26H1 and Windows Server 2025; any unpatched system on those versions is exposed to any local account holder. The flaw was fixed in Microsoft's record September 2026 Patch Tuesday (974 CVEs), was added to CISA's KEV on 2026-09-08 as one of two Windows zero-days reported as exploited in the wild, and has no known public PoC or confirmed ransomware use.
· Microsoft Windows 11 23H2, 24H2, 25H2, 26H1 · Microsoft Windows Server 2025 KEVmass
Use-After-Free in WebGL in Google Chrome Allows Code Execution Outside the Sandbox
CVE-2026-87464 is a use-after-free (CWE-416) in the WebGL component of Google Chrome affecting versions prior to 153.0.8010.36. A remote attacker can trigger it by luring a user to open a crafted HTML page, which corrupts memory in the browser's WebGL rendering path. Successful exploitation allows arbitrary code execution outside the browser sandbox, meaning the attacker escapes Chrome's process isolation and runs code with the privileges of the browser on the host. All Chrome users running an unpatched version before 153.0.8010.36 are affected. As of now there is no public proof-of-concept and this specific flaw is not in CISA KEV, although the Chrome 153 release headlines reference a separate V8 zero-day that was exploited in the wild.
· google chrome all versions prior to 153.0.8010.36mass
Entities are extracted by the model from each article. Watching an entity keeps it in this browser only (no account); the watchlist page and dashboard alerts use it.