News World That's Worth

Critical Android "SMS of Death" Vulnerability Exposes Remote Call Disruption and Bootloop Risks Security researchers have disclosed a severe, critical-severity vulnerability affecting Google’s Android operating system, dubbed the "SMS of Death" The zero-click exploit allows remote attackers to disrupt phone calls, sever emergency helpline connections, and potentially brick target devices into persistent bootloops through specially crafted text messages. Mechanics of the Exploit, Silent Delivery, and System Impact Stealthy Transmission and Zero UI Notification: Invisible to the Inbox : When the malicious message payload arrives, it bypasses standard SMS messaging applications entirely, generating no visual notification, chat history entry, or sender ID alert on the device screen. Active Call Termination : The payload immediately targets Android’s telephony and call-management framework, abruptly dropping active phone calls without warning. Critical emer...

GPT-6.1 Terra vs. Claude Opus 5.5: Benchmarking Benchmark Claims on Token Efficiency and Cost Performance The New Stack published a comprehensive real-world benchmark suite testing OpenAI  claims regarding its flagship model, GPT-6.1 Terra . OpenAI previously claimed that while Terra delivers slightly lower baseline accuracy performance than Anthropic’s Claude Opus 5.5 (a minor 5% variance), it operates 3x more token-efficiently and drops unit pricing by 75% , reducing overall task execution costs by roughly 90% (a 1/9 ratio). To evaluate these claims under real-world pressure, The New Stack put both models through three complex, multi-modal stress tests, setting both architectures to maximum effort mode ( max effort ). Benchmark Methodology, Practical Task Performance, and Cost Metrics Test 1: Full-Stack Code Debugging via Browser Visual Inspection: Task Overview : The models were tasked with diagnosing and repairing an incomplete software platform resembling a GitH...

Google Achieves Breakthrough in Project Suncatcher: Space-Borne TPU Runs Gemma 4 to Compose First Poem from Orbit Google has officially shared a major progress milestone for Project Suncatcher , its ambitious initiative aimed at deploying and operating specialized Tensor Processing Units (TPUs) in Earth's orbit. Following the successful launch of its experimental TPU-equipped satellite on October 1, 2026 , engineering teams spent the subsequent weeks conducting core diagnostic procedures, powering on onboard systems, and validating hardware integrity in the harsh vacuum of space. In-Orbit AI Inference, The Orbital Haiku, and Extreme Environment Stress Testing First Operational Inference Test and Orbital AI Poetry: The "Hello World" of Space Compute : Over the past week, Google transitioned from hardware diagnostics to active AI model execution. Engineering teams successfully loaded and executed the Gemma 4 open-weights model directly on the space-borne TPU node....

Meta Expands AI App Portfolio with Style Sketch and Waddle World Experimental Launches Meta has expanded its suite of experimental artificial intelligence applications, launching two new standalone experiences: fashion styling assistant Style Sketch and virtual pet companion Waddle World . These launches expand Meta’s growing portfolio of specialized AI-driven consumer applications, following the previous debuts of Pocket (an AI-powered game generation tool) and Vibes (a dedicated feed app for AI-generated short-form video content). App Features, Instagram Integration, and Regional Testing Rollout Style Sketch: Personalized AI Wardrobe and Fashion Recommendations: Instagram Profile Synchronization : Style Sketch connects directly with a user’s Instagram account to analyze existing photos, recognizing personal aesthetic preferences, color palettes, and individual wardrobe items. Users can also upload additional photos to refine their digital closet. Body-Type and Style Matc...

Binance Restores 90% of Stolen Assets Following $5.7 Billion Breach, Reopens Withdrawals on November 1 Binance has released an official operational update following the recent security breach that resulted in the unauthorized transfer of approximately $5.7 billion in digital assets. Demonstrating rapid financial recovery efforts, the exchange confirmed that 90% of all compromised funds have already been fully compensated using internal capital reserves. Additionally, Binance confirmed that user withdrawals officially reopen today, November 1, 2026 , adhering strictly to the security recovery schedule announced previously. Capital Mobilization, Top 10 Asset Recovery, and Liquidity Remediation Strategy Full Compensation Across Top 10 High-Market-Cap Assets: 100% Coverage for Core Tokens : Binance has completely restored token balances across its primary asset offerings. The fully reimbursed digital assets include Bitcoin (BTC) , Ethereum (ETH) , Tether (USDT) , BNB (BNB) , ...

Google Updates Gemini App with Native Document Generation and Inline PDF Editing Google has officially updated its Gemini application ecosystem, introducing built-in capabilities to generate, structure, and edit native documents, presentations, spreadsheets, and PDF files directly within the chat interface. The update eliminates the need to export raw AI outputs into secondary productivity applications, allowing users to execute real-time document modifications, visual conversions, and layout adjustments using intuitive inline prompts. Direct File Manipulation, Multimodal Input, and Platform Availability Multimodal Parsing and Interactive PDF Editing: Handwritten Notes to Structured PDFs : Demonstrating the system's multimodal understanding, Google showcased Gemini ingesting photos of handwritten lecture notes and automatically synthesizing them into a fully formatted, publication-ready study guide PDF. Contextual Drag-and-Highlight Edits : Users can directly interac...

OpenAI Closes $11 Billion Funding Round Below Target as Anthropic Surpasses Annualized Revenue Pace OpenAI has reportedly closed its latest private capital raise, securing $11 billion in new funding. However, according to sources familiar with the matter, the figure fell well short of the company’s ambitious $30 billion target. The capital injection brings OpenAI’s post-money enterprise valuation to $1.25 trillion  a landmark milestone, yet missing the firm's internal $1. 4 trillion valuation target as investor scrutiny intensifies over capital deployment efficiency and expanding burn rates across the generative AI sector. Capital Market Dynamics, Financial Metrics, and Anthropic Competition Investor Caution and Market Competition: Capital Crowding : Investors cite an increasingly crowded landscape as numerous technology firms simultaneously seek multi-billion-dollar funding rounds to capitalize AI infrastructure and foundation model R&D. Widening Losses vs. Top-Line G...

Sam Altman Sparks Backlash After Suggesting Personal PC RAM Be Redirected to Cloud AI Infrastructure In an interview on the Cheeky Pint podcast, OpenAI Chief Executive Officer Sam Altman ignited intense debate across the global hardware community after addressing the ongoing global semiconductor and memory supply crunch driven by enterprise artificial intelligence demands. When asked about the widespread RAM shortages and skyrocketing hardware prices affecting consumer PC builders, Altman offered a controversial perspective on global resource allocation, questioning the efficiency of high-performance memory sitting idle inside personal computers. The Controversial "Idle RAM" Arguments and Hardware Community Reactions Altman’s Stance on Resource Utilization: The Efficiency Dilemma : Altman described the current situation as "quite terrible, " pointing out that high-capacity RAM installed in personal computers worldwide remains idle for the vast majority of its...

Top AI Giants Google, Microsoft, OpenAI, and Anthropic Unite on Standardized In a major joint initiative aimed at respecting content creator consent and standardizing web-scraping rules, leading American artificial intelligence developers Google , Microsoft , OpenAI , and Anthropic have jointly pledged to honor a unified robots.txt instruction. Under the new agreement, all four platforms pledge to cease web crawling and training data extraction from websites that explicitly opt out via standardized site directives. Technical Specification, Developer Consensus, and Deployment Schedule Standardizing Web Crawling Directives via IETF Framework: Eliminating Webmaster Confusion : Previously, AI companies utilized fragmented user-agent tags and custom directives (such as GPTBot , Google-Extended , or ClaudeBot ), causing widespread confusion for webmasters attempt to block training bots. IETF Draft Specification Standard : The four tech giants agreed to implement a standardized ...

Spotify to Launch AI Song-Remixing Feature in Early 2027: Partners with Top Music Labels for Royalties and Chart Integration Spotify is preparing to introduce an interactive, artificial intelligence-powered song-remixing capability in the first quarter of 2027, with an official public unveiling targeted around the 69th Annual Grammy Awards in February. According to industry reports, Spotify is considering an add-on subscription tier priced between $4 to $5 per month for premium subscribers wishing to access the generative audio manipulation suite. AI Transformation Tools, Label Licensing Agreements, and Royalty Structure Creative Features and User Capabilities: Genre and Style Transformation : Premium subscribers will be able to leverage generative AI models to re-imagine existing tracks across different genres such as converting a traditional folk acoustic ballad into a high-tempo electronic dance music (EDM) track. Cross-Artist Vocal Covers : The feature allows users to genera...

Mistral Releases Flagship 1-Trillion Parameter Open-Weight Model "Mistral Large 4" (Le Chonk) French artificial intelligence startup Mistral AI has officially launched its latest flagship open-weight foundation model, Mistral Large 4 (codenamed Le Chonk ). Packing 1 trillion parameters , the open-weights release delivers state-of-the-art coding benchmark performance, exceptional token efficiency, and advanced cybersecurity defense capabilities designed for enterprise deployment. Benchmark Performance, Cyber Safety Capabilities, and API Pricing Structure Frontier Coding Capabilities and Benchmark Results: FrontierCode 2.0 Leaderboard : Mistral Large 4 achieved top-tier status as the highest-performing open-weight coding model, scoring 19.7% on the FrontierCode 2. 0 benchmark—surpassing DeepSeek V4. 1-Flash (19. 3%). Artificial Analysis Coding Agent Index : The model scored 48% , placing it closely behind GLM-5. 3. However, Mistral Large 4 accomplishes this perfor...