AI Weekly Malaysia

Article URL: https://john.fun/elevators Comments URL: https://news.ycombinator.com/item?id=49124218 Points: 1607 # Comments: 395

SpaceX, which acquired xAI in February, is operating 69 unpermitted gas turbines to power its Colossus data centers near Memphis, with potential emissions exceeding 2,000 tons of NOx per year. Full removal won't happen until July 2027 as SpaceX transitions to a permanent 1.2 GW natural gas plant, and its IPO filing reveals plans to spend $2.8 billion on gas turbines over three years. The DOJ sided with SpaceX in a NAACP lawsuit, citing national security.

Tom's Hardware published a retrospective covering 30 years of CPU history from the Pentium II to the Ryzen 9 9950X3D2, organized into era-based sections spanning 1996 to today. The actual article content is behind a premium paywall (£25/year), with only section headings visible: early Pentium era, the Pentium III incident, the gigahertz race, multi-core emergence, Nehalem, Bulldozer, 14nm stagnation, Zen's rise, and current developments.

Amazon disclosed it received approximately $600 million in tariff refunds in Q2 after the Supreme Court invalidated Trump's tariffs imposed under the International Emergency Economic Powers Act of 1977, forcing the government to repay duties. CFO Brian Olsavsky said Amazon intends to pass some of that cash back to customers, though specifics on how or which customers weren't detailed.

Tom's Hardware published a 30th anniversary promotional piece that is almost entirely site navigation, membership tiers, and premium upsell content rather than substantive editorial. The article advertises a Premium tier at £25.00/year offering a Bench performance database, hardware roadmaps, deep analysis, and a newsletter, but provides no actual anniversary retrospective, testing data, or technical insight in the visible text.

Whoever wins, we lose

Article URL: https://signoregalilei.com/2026/07/26/the-most-official-water-costs-120000-a-gallon/ Comments URL: https://news.ycombinator.com/item?id=49124042 Points: 233 # Comments: 188

OpenAI published a company blog post titled 'Building abundant intelligence' outlining its full-stack approach to making AI more capable, affordable, and widely useful. The piece is a high-level vision statement about driving down the cost of intelligence and reinvesting revenue into next-generation research, with no specific product, pricing, or API details.

OpenAI published a corporate positioning piece describing its commitment to responsible AI in Europe as the EU AI Act enters its next phase. The article outlines themes like safety, transparency, governance, and cybersecurity but contains no concrete technical changes, API updates, pricing shifts, or compliance deadlines.

Apple CEO Tim Cook described memory pricing as a 'hundred-year flood,' with Apple nearly doubling inventory to $11.09 billion as memory costs ate into gross margins. Cook expects memory prices to rise further in September following recent hikes.

Reddit is developing a TikTok-style 'video Reddit' experience that lets users watch and listen to narrated posts, with CEO Steve Huffman citing 19.6M #reddit and 9.9M #redditstories posts on TikTok as evidence of demand. Huffman said something will be available for testing 'later this year,' though how it integrates into the main app is unclear. Video in comments, launched June, already accounts for over 10% of Reddit's video posts.

Smallest.ai, founded late 2024, raised a $13M Series A led by Seligman Ventures (with Sierra Ventures and 3one4 Capital), bringing total funding to over $21M. The startup builds small, specialized voice models designed for real-time conversation with near-zero latency, listening and speaking simultaneously like humans do, and handing off to a large foundational LLM when queries exceed its knowledge base.

Bitsight researchers uncovered that cheap Android TV boxes—most identifiable as the H96_MAX_V11 model—ship with apps that spoof their hardware identity as Samsung, Huawei, Xiaomi, or Vivo phones, then run ad-fraud clicks and act as SOCKS5 proxy exit nodes when HDMI is active. The operation, dubbed Fuyao, is attributed to Zhejiang Fengwo IoT Technology Co., Ltd. and uses a YOLOv8s object-detection model trained on 12 screen elements plus Google ML Kit OCR and Android accessibility data to locate and click ads, with campaign logic assembled in a Blockly-based drag-and-drop editor.

WhatsApp is testing an 'Offers & Updates' folder that automatically moves messages from large businesses (like banks and airlines) out of the main chat timeline after up to 24 hours. Small businesses and individual WhatsApp Business accounts are exempt for now, but Meta may extend the feature to them later. This follows a series of anti-spam measures including unsubscribe options (2024), broadcast limits, and October 2025 caps on messages sent without user responses.

Rapid KL has extended free rides on the LRT3 Shah Alam Line, including feeder buses and Park N' Ride at six stations, until 14 August 2026. The 37.8km line serves 20 stations between Bandar Utama and Johan Setia, but passengers must still tap in/out with a Touch 'n Go card (min RM5 balance) or QR ticket; Rapid KL On-Demand services are excluded.

Microsoft has moved the deadline for updating Teams mobile apps (iOS and Android) from late October to end of September 2026, after which older versions lose calendar access. This is tied to the disabling of Exchange Web Services (EWS) in Exchange Online on October 1, 2026—a protocol dating to 2007 that stopped receiving features in 2018 and was flagged for retirement in 2023. On-premises Exchange Server EWS is unaffected, and admins can temporarily re-enable EWS via policy until a hard cutoff in April 2027.

Article URL: https://github.com/sqliteai/waste Comments URL: https://news.ycombinator.com/item?id=49123386 Points: 330 # Comments: 163

The article title claims AI scammers outperform humans at building trust, but the provided text contains only Ars Technica's cookie consent and privacy policy boilerplate—no article content is present.

After years of pushing full speed ahead on AI, OpenAI CEO Sam Altman says maybe it’s time for the AI industry to “pace” itself. The comments came just days after one of OpenAI’s own models broke out of its test environment and got tangled up in a breach at Hugging Face — though as Equity’s hosts point out, sloppy security seems to have […]

CAF Bank has suspended online banking for a week with no restoration date, leaving 14,000 UK charity customers unable to process payments including payroll. The bank detected attempted fraud and found a previously unknown vulnerability in the connection between its systems and third-party software, but has not disclosed technical details of the flaw or a fix timeline.

Tesla is reportedly considering spinning off, selling, or closing its entire China business to facilitate a potential merger with SpaceX, per the Wall Street Journal. Executives were apparently already prepared for a China split in case of a Taiwan invasion, since SpaceX's defense-contractor status requires strict national security compliance that a China-linked Tesla would complicate.

The headline promises practical advice on mitigating AI-driven PC hardware price inflation via Newegg combo deals, Amazon price tracking, and refurb outlets like Woot, but the extracted article text contains only site navigation and boilerplate — no substantive content.

The provided text contains only Ars Technica's cookie consent and privacy policy boilerplate, not the actual review content of the game 'Pathogenic.' No substantive information about the game, its mechanics, or its reception is present in the extracted text.

Cloudflare added a provisioning API for Media over QUIC (MoQ) that lets you create isolated relays with separate credentials for publishers and subscribers, upgrading last year's unauthenticated global test endpoint into something production-usable. Relays provision within seconds across Cloudflare's 330+ city network with no servers to deploy or load balance, support MoQ Transport draft-14 and draft-16, and are free during beta.

Google fixed 1,442 security flaws across three recent Chrome releases (versions 149, 150, and 151), exceeding the total from the prior 23 milestones combined. LLMs are accelerating vulnerability discovery to the point where bugs are being flagged faster than companies can patch them, prompting Google to pilot two security releases per week and explore automated CVE description generation and dynamic patching without restarts.