PM Modi Receives Uzbekistan's Highest State Honour During Two-Day Visit

Prime Minister Narendra Modi was conferred the Order of Oliy Darajali Dustlik, Uzbekistan's highest state honour, during a two-day official visit to the country. The award was presented by Uzbek President Shavkat Mirziyoyev in recognition of Modi's contributions to strengthening ties between the two nations. The honour underscores the growing strategic partnership between India and Uzbekistan. During the visit, both leaders held talks focused on expanding bilateral cooperation and aligning their respective development agendas.

Saudi Man Linked to 9/11 Hijackers Remembered as Friendly by Former Neighbor

Omar al-Bayoumi, a Saudi national, has been accused of helping two of the September 11 hijackers establish themselves in the United States in the months before the 2001 attacks. A woman who knew al-Bayoumi as a tenant recalled him as consistently cheerful and sociable, showing no signs of suspicious behavior. His alleged role in assisting the hijackers has made him a significant figure in investigations into the support network behind the 9/11 plot. The case highlights the ongoing efforts to uncover the full extent of foreign assistance that may have aided the attackers prior to that day.

Zambian Opposition Leader Charged with Treason After UN Building Hideout

Zambian opposition leader Mundubile has been formally charged with treason, authorities confirmed. Prior to the charges, he had sought refuge inside a United Nations building located in the country's capital. The move to hide at a UN facility drew significant attention to the political tensions in Zambia. The treason charge marks a serious escalation in the standoff between the opposition figure and the government.

Developer Builds AI Ransomware Response Agent Using TrueForge at 2026 Hackathon

A developer created AutoVault, an autonomous AI security agent, during the Agent Harness Hackathon 2026 to address the growing ransomware threat, which costs organizations over $20 billion annually. Built on TrueFoundry's open-source TrueForge harness, AutoVault can detect ransomware in 0.3 seconds and deploy five parallel subagents to investigate threats simultaneously. The system uses 23 MCP tools across four specialized servers and runs 20 Python scripts within a Daytona sandbox environment. To balance speed with safety, AutoVault incorporates a human-in-the-loop approval workflow that assigns risk scores from P1 to P4, requiring human sign-off before any irreversible actions are taken. The project aims to cut incident response times from hours to seconds while keeping human oversight intact for high-stakes decisions.

Anthropic CEO reassures industry as AI disruption fears grip software sector

Anthropic CEO Dario Amodei has sought to calm investor concerns, stating the company has no intention of undermining enterprise software firms. He described the growth of AI as a collaborative and positive development for the broader technology industry. Wall Street saw a temporary sell-off following a recent AI-related announcement, though software stocks have since recovered. Salesforce reported strong business growth and customer retention, reflecting deeper integration of AI into its offerings. Despite the reassurances, concerns around data security and enterprise autonomy in dealings with AI providers continue to linger.

How to sync React frontend and Node backend using a single RBAC policy

A common pitfall in web security is maintaining separate authorization logic on the frontend and backend, which causes rules to drift out of sync when changes are made. Role-Based Access Control (RBAC) can be implemented using a shared policy engine that maps roles to resource-action permissions, running identically on both the server and in the browser. On the backend, middleware intercepts requests and enforces the policy before any controller logic executes, making server-side enforcement the true security boundary. On the frontend, the same engine drives UI decisions — hiding or disabling controls — ensuring the interface never offers actions the server would reject. Using a single source of truth for authorization rules, including conditional checks like ownership validation, eliminates inconsistencies across both enforcement points.

Nepal Flash Flood Death Toll Rises to 734 as Rescue Operations Continue

The death toll from devastating flash floods in Nepal climbed to 734 as of Sunday. Rescue teams remained actively engaged in ongoing operations across affected areas. Assam Chief Minister Himanta Biswa Sarma reached out to eight Assam residents who were rescued in Nepal. The contact came as part of efforts to account for and assist Indian nationals caught in the disaster.

How OpenTelemetry Brings Full-Stack Observability to Modern Distributed Systems

Modern software built on microservices and serverless architectures offers scalability but creates significant operational complexity, making it difficult to diagnose failures across interconnected services. Without proper observability, engineering teams face prolonged incident resolution times, manual log correlation, and reactive rather than proactive problem detection. OpenTelemetry (OTel) addresses this by providing a vendor-agnostic, standardized framework to collect and export the three core telemetry signals: metrics, logs, and distributed traces. Its architecture typically involves instrumenting applications via OTel SDKs, routing data through a centralized collector for processing, and forwarding it to backend systems like Prometheus for metrics and specialized tracing tools. The approach aims to reduce mean time to resolution, improve developer productivity, and prevent performance issues from escalating into costly outages.

VeriTest Engineer Detects Suspiciously Clean Node Removal, Traces It to Hidden Bait Data

Late at night at the VeriTest office, security engineer Lena noticed a monitoring node had vanished from her probe overview with none of the usual warning signs of a real failure. Unlike normal decommissions, the node dropped from active to offline in a single beat, with clean logs and no retry alerts, prompting her suspicion. She traced the anomaly to a two-minute window around 2:39–2:47 AM, during which the same source profile appeared twice before the device went fully dark. Investigating further, she checked a private secondary observation network she had set up independently, where she found a mysterious single-line entry near the location of forged data she had previously planted as bait inside ACL's monitoring pipeline. The discovery suggests that an outside actor detected and responded to her planted data, confirming that the bait had drawn a deliberate, calculated reaction.

AWS Load Balancing Explained: When to Use ALB vs NLB in Cloud Architecture

Amazon Web Services offers two primary managed load balancers — the Application Load Balancer (ALB) and the Network Load Balancer (NLB) — each designed for distinct traffic handling needs. The ALB operates at OSI Layer 7, enabling content-aware routing based on HTTP headers, URL paths, and hostnames, making it suited for microservices, REST APIs, and web applications. The NLB functions at Layer 4, forwarding raw TCP/UDP packets without payload inspection, delivering sub-millisecond latency and instantaneous burst capacity for high-throughput workloads. A key architectural difference is that NLB provides static Elastic IPs per Availability Zone, while ALB relies on dynamic DNS resolution. Both load balancers support TLS handling and distribute traffic across EC2 instances, EKS pods, Lambda functions, and IP addresses across multiple Availability Zones.

Nepal floods revive Bihar’s annual nightmare: Why the state faces flooding every year

Security Report Flags High Flash Loan Risk on MEXC DeFi Platform Worth $5.24B

A DeFi security assessment dated August 30, 2026, identified five significant flash loan attack vectors targeting MEXC, a decentralized finance platform with a total value locked of approximately $5.24 billion across Ethereum and Layer 2 networks. The report, prepared by a senior DeFi security researcher, assigned MEXC an aggregate flash loan risk score of 7.4 out of 10, placing it in the 'High-Risk – Immediate Mitigation Required' tier. The most critical vulnerabilities include oracle price-feed manipulation on L2 AMM pools, a re-entrancy flaw in the liquidity mining reward contract, and a cross-chain bridge race condition, with potential losses estimated at up to $300 million per vector. Additional risks involve governance token oracle manipulation and a collateral-ratio bypass in the lending module that could trigger under-collateralised liquidations worth $50–$100 million. The report recommends immediate remediation across the affected smart contracts, including MEXCPerpetual.sol, MEXCOracle.sol, MEXCRewardDistributor.sol, and related modules.

Scientists Produce Smallest-Ever Big Bang to Probe Early Universe Conditions

Scientists have successfully created the smallest version of a Big Bang ever achieved in a laboratory setting. The experiment was designed to study the extreme state of matter that existed in the early universe. The findings challenge previous assumptions about the minimum size of atoms required to produce such conditions. This discovery redefines the threshold at which atomic collisions can generate the dense, hot matter believed to have existed shortly after the universe's birth.

Meta's US Teen Safety Rules Spark Debate Over Age Verification in India

Meta has agreed to strict usage limits for users under 18 in the United States, including a two-hour daily screen time cap, a midnight-to-6am access block, and muted notifications during school hours. These measures have reignited discussions in India about the need for similar child safety protections on social media platforms. Experts in India are pointing to the agreement as evidence that stronger age verification mechanisms are essential to enforcing such rules effectively. The debate highlights a growing concern among child safety advocates about the absence of robust regulatory frameworks governing minors' social media use in India.

H-4 EAD: What latest US move means for Indian spouses of H-1Bs

Arbitrary code execution in QubesOS via copy-to-VM error reporting backchannel

Article URL: https://www.qubes-os.org/news/2026/08/29/qsb-118/ Comments URL: https://news.ycombinator.com/item?id=49496918 Points: 7 # Comments: 0

Trump wants 'Lake America'. Canada puts up a 24-foot sign saying otherwise

Inside the journey to build a fully indigenous AK-203

DeepMind's WeatherNext AI Delivers 24 Extra Hours of Cyclone Warning

Google DeepMind published research in Nature on August 6, 2026, introducing WeatherNext Cyclones, an AI model that predicts a tropical cyclone's track, intensity, and wind structure within a single unified system. Trained on nearly 20 terabytes of atmospheric data and records of close to 5,000 historical storms, the model can generate up to 1,000 forecast scenarios in under a minute on a single processor. Benchmarks against 2023–2025 storms showed it provided roughly 24 hours more reliable warning than leading operational systems, with 28% better track accuracy and 6% better intensity accuracy when blended with official forecasts. The model was used operationally by the U.S. National Hurricane Center during the 2025 Atlantic hurricane season to help forecast Hurricane Melissa's rapid intensification and Jamaica landfall. DeepMind has open-sourced the code and pretrained weights, making the system freely available to researchers worldwide.

Noida Airport’s big takeoff faces airline turbulence