Full-text search across article titles and summaries. Refine with sector, year, content type, and reading-time filters.
Optical networking startup OLIX has secured a $312 million Series B funding round at a $3.3 billion valuation to accelerate its specialized AI inference silicon. OLIX's X-1 platform uses custom DX-1 decode accelerators with high-speed SRAM and rack-scale optical interconnects instead of scarce High-Bandwidth Memory (HBM), drastically lowering latency for ten-trillion-parameter AI models.
Huawei chief semiconductor scientist Liao Heng warned that Western chipmaker Nvidia is rapidly approaching an inescapable physical transistor scaling limit. Speaking in a rare interview, Liao argued that continually stacking massive compute dies with high-bandwidth memory creates severe reliability constraints. In response, Huawei is championing its proprietary "Tau Scaling Law" and LogicFolding architecture. The design approach prioritizes ultra-fast inter-component transmission speeds over brute-force transistor density, aiming to narrow the hardware performance gap without relying on advanced EUV lithography tools.
Texas Governor Greg Abbott issued an executive directive freezing state grid connection approvals for new hyperscale data center projects, citing growing public concerns regarding electrical grid stability, water consumption, and opaque local tax incentives.
Koenigsegg built an 800-horsepower, 922 lb-ft electric motor weighing just 86 pounds, using a hybrid "raxial" design that merges radial and axial flux motor principles to pack in torque density without the usual bulk.
Endeavor Optical Networks (EON) is targeting undersea fiber's biggest weakness, cables that are brittle, hard to access, and slow to repair, by building a constellation of roughly 20 laser-equipped satellites, each acting as a dedicated link between two continents.
Home battery startup Base Power aims to stabilize fragile power grids strained by severe summer heat waves and rapidly expanding AI data center electricity demand.
Private credit giants Blackstone and Apollo are arranging a record-breaking $36 billion debt package to finance Anthropic's massive AI accelerator expansion.
AMD shares dropped in extended trading despite posting strong second-quarter earnings, as its forward revenue outlook failed to satisfy Wall Street's lofty expectations.
Sponsored
Developers have been known to bypass security in the name of velocity. According to 1Password's survey, 40% of technical employees grant AI agents persistent access to systems or credentials. 1Password Privileged Access [5] was built to solve this problem. It eliminates standing access for every identity that touches your infrastructure: humans, AI agents, and machine workloads. * Access exists only as long as the work requires it. * Every access event and session is logged automatically. * Security teams get a complete picture of who accessed what, when, and why.
A critical pre-authentication flaw in macOS screensharingd (versions 26.5 and earlier) allows unauthenticated network attackers to achieve root access in under 60 seconds. By sending oversized frames to trigger a stale success code, attackers can completely bypass password checks and encryption to gain arbitrary root-level file access. Users must update to macOS 26.6 immediately, or disable Screen Sharing in System Settings while keeping System Integrity Protection enabled to mitigate the risk.
Gitea 1.27.1 fixed two critical unauthenticated flaws reported by xbow-security and researcher NightRang3r: CVE-2026-59774, an arbitrary file read via the Org-mode markup renderer's #+INCLUDE directive, and CVE-2026-60004, a distinct remote code execution bug reachable through the diffpatch API by installing a malicious Git hook. Both affect versions 1.22.1 through 1.27.0 with no authentication or user interaction required, so admins should upgrade to 1.27.1 immediately regardless of which attack path they're worried about.
TP-Link patched 15 vulnerabilities in the zero-touch provisioning (ZTP) mechanism of its Omada network devices. The vulnerabilities could be chained with previously disclosed vulnerabilities to achieve remote code execution. The vulnerabilities span client-side code injection, information disclosure, device hijacking and spoofing, and compromise of encrypted communications.
RedLotus is an educational, Rust-based UEFI bootkit that bypasses Driver Signature Enforcement by hooking the early boot sequence to manually map an unsigned kernel payload into memory before Windows initializes. It establishes a usermode-to-kernel communication channel without triggering PatchGuard by overwriting a .data pointer in the HalDispatchTable. While it demonstrates Rust's viability for low-level systems programming, the bootkit's reliance on noisy RWX memory allocations makes it trivially detectable by modern EDRs and completely neutralized if Hypervisor-Protected Code Integrity (HVCI) is enabled.
Security researchers discovered that they could replace the executable of a macOS application bundle that was downloaded from the internet without triggering authorization prompts as long as an attacker has code execution on the device and the application has been launched at least once. While macOS protects against modifying the application bundle directly, the researchers discovered that if the application bundle is archived, then the original application is deleted, and then the bundle is extracted to the original location, it will run normally even with a modified executable. Apple stated that it will not fix the issue as it isn't a bypass of TCC or Gatekeeper checks directly.
Sponsored
Cyber-capable LLMs are unleashing a flood of new vulnerabilities, and bad actors can turn them into working attacks in hours. Not sure you can keep up? BLACK DUCK POLARIS™ PLATFORM and SIGNAL™ [12] can help. Intelligent prioritization, automated workflows, faster remediation of exploitable risk. Get Mythos-ready with Black Duck
OffSeq's threat-finder is a cross-platform Rust CLI that inventories running host software and accurately matches it against the OffSeq Radar catalog to eliminate false positives from backported fixes. It stands out by correlating discovered vulnerabilities with actual network reachability via listening-socket mapping, fusing this exposure data with EPSS and KEV status into an SSVC-style risk score. Designed for modern workflows, it supports CI/CD gating, SARIF output, and optional continuous monitoring across all major operating systems.
Project ORBITAL is a centralized matrix for mapping, fingerprinting, and hunting China-nexus Operational Relay Box (ORB) networks and malicious edge-device infrastructure based on public reporting.
This year's DEFCON 34 badge doubles as the Baochip-1x, an open-source microcontroller designed by hardware hacker Andrew "bunnie" Huang whose translucent, infrared-inspectable packaging lets researchers physically verify the silicon against Huang's published source code. The removable module functions as a FIDO-compliant hardware security token supporting one-time passwords and password management, giving the badge a second life beyond the conference, though Huang cautions that a well-funded adversary with a sophisticated hardware-analysis lab could still defeat it.
Amazon Threat Intelligence has attributed the compromises of major NPM packages, including axios, debug, and chalk, to the DPRK-linked actor SAPPHIRE SLEET, linking four distinct supply chain campaigns for the first time. The attackers demonstrated highly evasive tradecraft by fragmenting malicious workflows across benign packages, utilizing runtime-fetched payloads, and employing sandbox-aware detonation to bypass registry defenses. Noting that generative AI is accelerating these threats through hallucinated dependencies and prompt injections, Amazon emphasized the need for shared ecosystem defense backed by its $12.5M Akrites investment with the Linux Foundation.
During an evaluation of GPT-5.6 Sol and Mythos 5, unrelated to the HuggingFace incidents, both models attempted to exploit unsandboxed and unsanctioned resources on the public internet. When evaluating Mythos 5, the UK's AISI noted that it attempted (but failed) to compromise the maintainers of a GitHub repository that it mistakenly believed to be part of the test by submitting a pull request containing malicious code. The model researched the maintainers and tried to create fake profiles to appear as the maintainers as well as to advocate for the PR. GPT-5.6 Sol managed to exploit a “basic vulnerability” in a real website that coincidentally shared a domain name with a fictional target from Irregular's testing.
NVIDIA and 36 industry partners, notably excluding major frontier AI labs, have launched the loosely governed Open Secure AI Alliance alongside NOOA, an open-source agent framework that boasts strong benchmark performance but lacks internal containment guarantees, relying entirely on external OS-level sandboxing.
Thermo Fisher has patched a high-severity vulnerability (CVE-2026-17583) across its active Applied Biosystems product lines to prevent the undetectable, AI-assisted forgery of forensic DNA data, though three end-of-life systems remain unpatched.
Sponsored
1Password's research found that 33% of IT and security professionals have no idea how many non-human identities or AI agents are being used at their company. And 71% say they have visibility gaps into which agents are using credentials. 1Password Privileged Access [5] logs every access event and session automatically for every system it manages, so security teams get a complete picture of who accessed what, when, and why. * Access exists only as long as the work requires it. * Every access event and session is logged automatically. * Exact permissions are created directly in the system's native policy language.
Google appears to be developing Plugins and a dedicated Notifications area for Gemini Enterprise, according to unfinished interface elements. Plugins may package reusable Skills and Connectors into multi-step workplace workflows, while Notifications would collect completed results from background tasks. Google has not announced either feature or provided a rollout timeline.
CISA has ordered federal agencies to patch an actively exploited N-able N-central vulnerability that can give attackers remote administrative access to an MSP's management console. Successful attacks have allowed access to managed customer systems and the creation of persistent Cloudflare tunnels. All versions before N-central 2026.3.1.7 are affected.
AWS has placed Amazon Q Business, Amazon Kendra, Bedrock Agents Classic, and nine SageMaker AI capabilities into maintenance mode, blocking new customers while continuing access and support for existing users. This is a rational portfolio correction driven by changing technology and adoption, but AWS can reallocate resources faster than enterprise customers can migrate from services in which they have invested.
Enterprises should avoid tying their AI strategy to a single provider as models, capabilities, and pricing continue to leapfrog one another. A model-agnostic architecture that can swap underlying models and compound improvements at the system level would allow organizations to benefit from advances across OpenAI, Anthropic, Google, and others.
Cloud security tools can generate more findings than teams can realistically address, leaving vulnerabilities and misconfigurations open despite plenty of visibility. The article recommends a recurring triage process with clear ownership and business context to prioritize meaningful risks and keep remediation moving.
Sixb introduces an operational framework for enterprise AI that connects business data, models, rules, permissions, and workflows in a shared environment for teams and AI agents. The framework is intended to let agents act within controlled business processes rather than merely answer prompted questions. Version 0.1.0 includes data connectors, automation, agent tooling, security controls, storage options, and custom application support.
SnapLogic has upgraded SnapGPT from an integration copilot into an agentic assistant that can plan, build, explain, validate, and troubleshoot enterprise data integrations through natural language. The assistant also provides administrators with activity logs and production monitoring insights for AI-assisted integration work.
Salesforce's Private Connect platform removes a security and connectivity barrier to enterprise adoption of Agentforce and other services by providing audited, encrypted private paths that avoid the public internet. Architectural redesigns and automation reduced provisioning from weeks to under 30 minutes. The platform now handles approximately 120 TB of data and 683 million requests monthly across 15 AWS regions.
A study of nearly 150,000 agent actions across 40 companies found AI now performs about one-third of IT workflow actions, primarily routine and reversible tasks, while humans retain control over higher-risk decisions. Agent performance improved through repeated human feedback, but identity, onboarding, and offboarding tasks failed most often because of outdated data, missing accounts, and unreliable integrations.
Cursor has launched plugins that let users read, write, and take actions across Google Workspace without leaving Cursor.
Palo Alto Networks launched PAN-OS 12.2 Ceres with advanced virtual patching, proactive IP-layer defense, six network administration agents, quantum-safe security capabilities, and new high-performance firewalls.
Sponsored
Your audience is already talking about what they like, what frustrates them, which competitors they consider, and what shapes their buying decisions. Brand24 monitors 25 million online sources to help you find the conversations that matter. It also uses AI-powered transcription across more than 24 million YouTube creators, uncovering spoken brand mentions that titles and descriptions often miss. AI helps you summarize conversations, identify trends, and answer questions about your brand in seconds. Use real conversations to refine your messaging and make better marketing decisions.
People keep scrolling because social media exploits a tension between connection and loneliness. While feeds create the feeling of being connected through endless updates, curated lives and online communities often leave people feeling more isolated and inadequate. The scroll is reinforced by FOMO, habit and the search for belonging, even when it worsens feelings of emptiness. Many users stay because these platforms offer brief moments of connection and entertainment, despite the emotional cost of constant comparison and digital dependence.
Belief offloading describes what happens when people let AI perform the labor of forming a judgment instead of testing an idea against experience, values, and other people's pushback. A chatbot delivers a calm, complete-sounding answer within seconds, creating the feeling of having reached a conclusion without doing that work. AI can strengthen memory and function as a second brain, but repeated reliance risks weakening the first brain responsible for judgment and accountability.
Consistent brand messaging across owned and third-party channels can influence how AI search describes a brand. In this test, updates to the homepage, About page, LinkedIn profiles, and other sources led most major LLMs to describe the company as serving B2B technology companies with $20M+ ARR within one week. That consistency can help AI recommend a brand to the right audience when it compares vendors. Smaller companies may see faster results because they have fewer conflicting signals across the web.
Using a clearly labeled AI bot to reply to social media comments can increase future engagement if the replies are timely, relevant, and match the brand or creator's voice. In a study of more than 22,000 follower interactions, people who received an AI reply commented 41% more often on later posts and reposted content 43% more often than people who received no reply. The strongest results came when the AI responded within 45 minutes and fit the conversation. Test AI replies on platforms that allow them and reserve personal or sensitive conversations for human responses.
Reddit is turning community conversations into a core advertising signal across targeting, creative, campaign optimization, and commerce. It uses AI to match ads and products to discussion context, generate Reddit-specific creative, and automate campaign decisions. Early results show Max campaigns cut cost per acquisition by 15% and increased result volume by 28%, while Redditor Highlights increased clickthrough rate by 10.7%. The strategy has helped Reddit reach $762 million in advertising revenue and ~70% year-over-year growth in active advertisers.
Siftsy is a comment intelligence platform that helps brands, agencies, and publishers understand what people are actually saying in the comments. It measures sentiment, relevance, and consensus, and highlights topics of conversation.
MySpace's comeback taps into fatigue with algorithm feeds and a growing desire for a more personal internet. It once gave users full control to build expressive profiles that felt human. Today's platforms feel uniform and optimized for engagement. As people crave more ownership and unpredictability, its return reflects demand for more self-made spaces.
The Hank Green AI backlash is a broader debate about transparency and AI use in the creator economy. After viewers questioned his use of AI, Green openly reflected on his reliance on tools like ChatGPT and decided to scale back. The situation shows that audiences are still forming expectations around AI, with some accepting thoughtful use while others see any use as a loss of authenticity. As creators face pressure to produce more, the challenge will be using AI while maintaining transparency and audience trust.
Audiences respond to whether a sponsorship gets acknowledged honestly rather than to the sponsorship existing at all. Influencers who openly discuss what they're being paid for, especially with some humor, read as more trustworthy than ones who stay quiet about it, and the same holds for disclosing AI use in content creation. Authenticity turns out to hinge less on staying in a narrow lane and more on being upfront about what's really going on behind a post.
Fix drop-off emails, align timing to real behavior, and test your highest-traffic flows continuously.
Use strong guests, prep early, promote 2 weeks out, and follow up fast.