OpenAI Korea GM Harrison Kim revealed the company has made real progress with Samsung Electronics on joint production and research for its next-generation chips, expanding a relationship that so far has been mostly about memory supply and enterprise software. OpenAI's current inference ASIC, Jalapeño, was designed with Broadcom and built at TSMC in under 18 months and is already in mass production there, so any Samsung involvement more plausibly targets its successor, which is reportedly approaching tapeout.
TSMC posted NT$514.8 billion ($16.1 billion) in August revenue, up 53.3% year-over-year and 10.1% from July, hitting a fresh monthly record as AI-driven chip demand keeps outpacing supply. It came alongside a separate announcement with Dutch equipment maker ASML: TSMC plans to bring ASML's High-NA EUV lithography into large-scale manufacturing starting in 2030 for its most advanced nodes, as AI accelerators push toward increasingly complex transistor architectures.
NVIDIA has announced CUDA Rust and introduced two specialized compilation tracks that allow developers to author and optimize GPU compute kernels directly in Rust. This release brings Rust's strong compile-time memory safety guarantees and modern tooling ecosystem to GPU-accelerated workloads. Developers implementing embedded vision, intensive sensor fusion, and real-time state estimation on mobile platforms can now exploit GPU compute power while minimizing memory-related crashes and runtime concurrency bugs.
AMD and Embedded LLM teams took MiniMax M3's day-0 vLLM implementation from 109.1 to 342.4 output tokens/s/GPU on AMD's Instinct MI355X, a 3.14x gain, by repeatedly asking five questions as the bottleneck kept shifting: what local shape reached each rank after tensor-parallel sharding, what work repeats every layer, which bytes actually move, whether a claimed fast path actually executed, and which queue grows once kernels stop being the limiter. Big wins came from fusing the model's shared expert into the same grouped GEMMs as routed experts (up to 30% faster at low concurrency), and reusing sparse-attention block selections across adjacent layers instead of recomputing per layer.
Besxar, a $9.35 million seed-funded startup, flew its first mission on July 5 aboard a SpaceX Falcon 9 alongside a Starlink batch, carrying its own semiconductor wafers plus test wafers from the University of Virginia and UT Austin. The company's pitch is simple: Earth-based fabs spend billions on pumps to create the vacuum semiconductors need, but in space vacuum is free, so Besxar wants to manufacture higher-quality, lower-cost wafers in orbit and return them to Earth to sell.
Google plans to invest at least $15 billion in Finnish AI infrastructure through 2028, its largest single investment in Europe, building on over 15 years of presence there. Alongside the buildout, Google signed a 22-year power agreement with Finnish utility Fortum and said it will explore business models that could support new nuclear reactors at Fortum's Loviisa site, tying the compute expansion directly to a long-term power source.
Gamers Nexus published a two-hour investigation, done with security researchers MrBruh and uturn using packet capture and firmware analysis, alleging LG's smart TVs constantly log and upload user data even offline or in standby, scan local Wi-Fi networks for other devices, and record and store ambient audio through the TV's microphone for later retrieval, even when the TV appears off. LG called the claims untrue, saying its TVs only process voice data when the remote's voice button is pressed or its "Hi LG" wake word is detected after a user manually enables far-field voice recognition, with wake-word audio processed entirely on-device and deleted immediately if no match is found.
XPENG commissioned its humanoid robot production lines, and its IRON robot autonomously walked off them after completing assembly, a step the company frames as moving from R&D prototypes to real line manufacturing.
Chinese AI chipmakers, including Huawei and Cambricon, have sharply raised prices for current and next-gen processors.
Anthem is a long-range precision strike weapon aimed at a gap Covenant argues current Western arsenals lack.
Analog Devices agreed to buy Alif Semiconductor for $1.35 billion cash plus up to $200 million contingent on performance, adding microcontrollers built to run neural networks directly on-device rather than sending data to a data center.
JD Logistics plans to procure 3 million robots, 1 million autonomous vehicles, and 100,000 delivery drones over five years to automate picking, sorting, transport, and last-mile delivery.
Vention opened a Montreal lab focused on robotic manipulation, industrial data collection, and post-training robotics foundation models using learning-from-demonstration, reinforcement learning, vision models, and motion planning.
New CEO John Ternus led his first keynote, headlined by the iPhone Duo, Apple's first foldable, a passport-style device with a 5.4-inch outer display and a 7.6-inch inner display that automatically reflows apps based on how the phone is folded, starting at $1,999. Apple also unveiled the iPhone 18 Pro and Pro Max ($1,199/$1,299, up $100 from last year) with new A20 Pro processors, a variable mechanical aperture camera, and 24-30 hour battery life, alongside the Apple Watch Series 12 and Ultra 4, which add a 5-second heart rate sensor Apple claims is more accurate than a chest strap.
OpenAI Korea GM Harrison Kim revealed the company has made real progress with Samsung Electronics on joint production and research for its next-generation chips, expanding a relationship that so far has been mostly about memory supply and enterprise software. OpenAI's current inference ASIC, Jalapeño, was designed with Broadcom and built at TSMC in under 18 months and is already in mass production there, so any Samsung involvement more plausibly targets its successor, which is reportedly approaching tapeout.
TSMC posted NT$514.8 billion ($16.1 billion) in August revenue, up 53.3% year-over-year and 10.1% from July, hitting a fresh monthly record as AI-driven chip demand keeps outpacing supply. It came alongside a separate announcement with Dutch equipment maker ASML: TSMC plans to bring ASML's High-NA EUV lithography into large-scale manufacturing starting in 2030 for its most advanced nodes, as AI accelerators push toward increasingly complex transistor architectures.
NVIDIA has announced CUDA Rust and introduced two specialized compilation tracks that allow developers to author and optimize GPU compute kernels directly in Rust. This release brings Rust's strong compile-time memory safety guarantees and modern tooling ecosystem to GPU-accelerated workloads. Developers implementing embedded vision, intensive sensor fusion, and real-time state estimation on mobile platforms can now exploit GPU compute power while minimizing memory-related crashes and runtime concurrency bugs.
AMD and Embedded LLM teams took MiniMax M3's day-0 vLLM implementation from 109.1 to 342.4 output tokens/s/GPU on AMD's Instinct MI355X, a 3.14x gain, by repeatedly asking five questions as the bottleneck kept shifting: what local shape reached each rank after tensor-parallel sharding, what work repeats every layer, which bytes actually move, whether a claimed fast path actually executed, and which queue grows once kernels stop being the limiter. Big wins came from fusing the model's shared expert into the same grouped GEMMs as routed experts (up to 30% faster at low concurrency), and reusing sparse-attention block selections across adjacent layers instead of recomputing per layer.
Besxar, a $9.35 million seed-funded startup, flew its first mission on July 5 aboard a SpaceX Falcon 9 alongside a Starlink batch, carrying its own semiconductor wafers plus test wafers from the University of Virginia and UT Austin. The company's pitch is simple: Earth-based fabs spend billions on pumps to create the vacuum semiconductors need, but in space vacuum is free, so Besxar wants to manufacture higher-quality, lower-cost wafers in orbit and return them to Earth to sell.
Google plans to invest at least $15 billion in Finnish AI infrastructure through 2028, its largest single investment in Europe, building on over 15 years of presence there. Alongside the buildout, Google signed a 22-year power agreement with Finnish utility Fortum and said it will explore business models that could support new nuclear reactors at Fortum's Loviisa site, tying the compute expansion directly to a long-term power source.
Gamers Nexus published a two-hour investigation, done with security researchers MrBruh and uturn using packet capture and firmware analysis, alleging LG's smart TVs constantly log and upload user data even offline or in standby, scan local Wi-Fi networks for other devices, and record and store ambient audio through the TV's microphone for later retrieval, even when the TV appears off. LG called the claims untrue, saying its TVs only process voice data when the remote's voice button is pressed or its "Hi LG" wake word is detected after a user manually enables far-field voice recognition, with wake-word audio processed entirely on-device and deleted immediately if no match is found.
XPENG commissioned its humanoid robot production lines, and its IRON robot autonomously walked off them after completing assembly, a step the company frames as moving from R&D prototypes to real line manufacturing.
Chinese AI chipmakers, including Huawei and Cambricon, have sharply raised prices for current and next-gen processors.
Anthem is a long-range precision strike weapon aimed at a gap Covenant argues current Western arsenals lack.
Analog Devices agreed to buy Alif Semiconductor for $1.35 billion cash plus up to $200 million contingent on performance, adding microcontrollers built to run neural networks directly on-device rather than sending data to a data center.
JD Logistics plans to procure 3 million robots, 1 million autonomous vehicles, and 100,000 delivery drones over five years to automate picking, sorting, transport, and last-mile delivery.
Vention opened a Montreal lab focused on robotic manipulation, industrial data collection, and post-training robotics foundation models using learning-from-demonstration, reinforcement learning, vision models, and motion planning.
Anthropic Alignment Science Lead Evan Hubinger said he personally believes there is a greater than 10% chance AI could “kill all humans” within the next decade. He added that Anthropic still does not have a clear solution for aligning superintelligent AI, following another researcher's resignation over the pace of development.
2026.09.11·cbsnews.com·4 min read·TLDR IT Oracle's cloud infrastructure revenue grew 121% year over year as demand for AI compute continued to surge. The company spent $28.5 billion on capital expenditures during the quarter and maintained plans for $90–95 billion in spending this fiscal year
2026.09.11·businessinsider.com·3 min read·TLDR IT Microsoft has launched a public preview of Dynamics 365 Activate, an AI-powered tool to help partners and customers migrate from Salesforce to Dynamics 365 with less manual effort and lower risk. The company says the tool profiles data, relationships, and customizations to create a migration blueprint, and plans additional CRM and ERP migration scenarios later this year.
2026.09.11·theregister.com·2 min read·TLDR IT AI has shifted left from code repositories to coding agents on developer endpoints with broad privileges, where AppSec has limited control. AI-driven discovery is also flooding defenders with vulnerabilities. The better approach is to eliminate entire vulnerability classes rather than fix tickets one by one, while rethinking human-in-the-loop oversight when agents can both write and remediate code.
2026.09.11·resilientcyber.io·13 min read·TLDR IT OpenAI nonprofit board member Paul Christiano says the industry is not currently on track to prevent a potentially catastrophic loss of control over advanced AI. His warning adds to growing concern around how quickly increasingly autonomous systems are being developed.
2026.09.11·theguardian.com·5 min read·TLDR IT Google is rolling out five new agentic Gemini capabilities across Workspace that complete cross-app tasks in the background without switching tabs. Gemini can create Slides from Chat, build Sheets from Drive, draft and send emails from Docs, turn Gmail threads into structured Docs briefs, and convert Docs into branded decks. The features are rolling out now to Business, Enterprise, Google AI Pro and Ultra, and education plans.
2026.09.11·workspace.google.com·5 min read·TLDR IT Accenture and Google Cloud launched a new Gemini Enterprise group, including a 1,000-person FDE workforce, to scale agentic AI adoption. The group will help enterprises move from AI pilots to production with industry-specific solutions and implementation support. YouTube saw 11% higher customer sentiment and 37% lower handle time using Gemini Enterprise.
2026.09.11·newsroom.accenture.com·2 min read·TLDR IT Latham & Watkins has bought Nvidia GPU servers and is running customized open-weight models on infrastructure it controls. The move shows large enterprises exploring private AI stacks for greater control over sensitive data, costs, and model choice.
2026.09.11·ft.com·4 min read·TLDR IT Mistral says it helped a European energy operator migrate 40,000 lines of Fortran 77 reservoir-simulator code to C++ using supervised AI-agent workflows. The team built a numerical parity harness first, documented the codebase with over a hundred agents, and found that structured workflows with human review gates beat full autonomy. The first sprint covered core functionality of a 300,000-line system.
2026.09.11·mistral.ai·6 min read·TLDR IT The article compares Keycloak's all-in-one identity server with Ory's modular, headless services, arguing the choice hinges on who owns the login experience and operational load.
2026.09.11·cerbos.dev·11 min read·TLDR IT Nvidia is combining Nemotron models with Palantir's platform to automate supply-chain decisions while keeping proprietary enterprise data under tighter control.
2026.09.11·nvidianews.nvidia.com·2 min read·TLDR IT Sponsored
Ahrefs Brand Radar [5] tracks your brand across 471M+ real search-backed prompts.
* EVERY MAJOR PLATFORM – ChatGPT, AI Overviews, AI Mode, Gemini, Perplexity, Copilot * BENCHMARK COMPETITORS side by side, on any domain * SEE WHAT AI CITES – the pages, videos, and Reddit threads behind each answer * CUSTOM PROMPTS – add your buyers' exact questions, checked daily
Sponsored YouTube videos can extend their marketing value into AI search when the video is cited, especially through the description text that names the sponsor. The sponsor appeared in 40% of AI answers that cited a sponsored video, compared with 32.3% when the video was not cited, and the lift reached 1.35x on unbranded prompts. Brands should make specific description language part of influencer contracts and track YouTube citations for the prompts that matter to them, rather than treating sponsorship as a view-based channel alone. The opportunity is temporary because YouTube citation rates fell 43% within a month, so citation monitoring needs to be ongoing.
Shopping ad CTRs are rising while impressions are falling across thousands of campaigns, with clicks relatively flat or slightly down. One hypothesis is that Google's AI Overviews are appearing on queries less likely to generate ad clicks, leaving Shopping ads for higher-intent searches. The data doesn't prove AI Overviews are causing the decline, but advertisers should look beyond CTR and track impressions, clicks, and overall traffic to understand whether performance is actually improving.
Increase sales by labeling a mid-priced product as “recommended” instead of your most expensive option. In eight experiments involving more than 6,200 participants, sales of a recommended cafe drink increased 58% when a pricier option was added, while sales were unchanged without the recommendation. Consumers appear more receptive to recommendations when the suggested product is cheaper because the brand seems less motivated by profit. Use objective badges such as “Bestseller” when recommending a premium product, and test this pricing tactic on comparable products where customers can easily evaluate the price difference.
Sometimes a brand's old identity becomes valuable again simply because the category around it moved. As dining shifted toward fast casual, delivery, and pricier meals, Chili's leaned into value, booths, and the Triple Dipper, and a format that once looked dated became the countertrend against expensive takeout. Barnes & Noble made a similar bet, handing individual stores control over what to stock and display instead of running every location identically, so each one behaves like a local bookstore again. Both bets worked because the brand stopped trying to outrun its category and let the category's shift make its old strengths relevant.
Sponsored
KRITHIKA S., Stripe and OpenAI's first marketing hire and now head of marketing at Thrive Capital, joins CHRISTY ROACH (CMO, AirOps) live on Sep 17. They break down a new study of 300+ marketing teams: how high-growth teams hire, budget, and win more resources. Free, one hour.
Google released an open-source tool that tests whether an ad actually caused a sale, not just which platform claims credit for one. GeoX works by comparing regions where a campaign keeps running against regions where it's paused or given more budget, then measuring how sales in each group differ from what historical data would have predicted anyway. It can track spend across platforms beyond Google's own and combine online and offline sales in the same test. It plugs into Google's existing Meridian measurement software, so the results can also refine how a budget gets split across channels.
Amazon has spent months locking AI shopping agents like Claude and Gemini out of its own storefront while building its own AI shopping tools, then quietly became ChatGPT's top retail advertiser between April and August instead. It's now opening that same access to other brands that already buy through its demand-side platform, letting them run ads inside ChatGPT conversations too. Campaigns are still set up and managed inside Amazon's own systems, while OpenAI's ad system keeps control over which ad actually gets served.
Time magazine is testing ads designed specifically for AI agents, exposing them to brand-verified facts before they access webpages. The goal is to influence what AI systems retain and potentially correct inaccurate information about brands. As agentic traffic grows, companies are also optimizing websites and product data for AI discovery and recommendations. Brands are beginning to treat AI agents as an emerging advertising and distribution channel.
Skincare spent decades built around the idea that a good product should be invisible. Starface inverted that by making a pimple patch impossible to miss, then expanded through Target, Walmart, Sephora, and Ulta specifically because its bright yellow stars stand out against a retail aisle of clinical whites and pastels. That same bet is now showing up industry-wide, with LED light masks and other visible skincare wearables replacing creams that disappear on the skin.