Models, products and AI startups, near-duplicates collapsed to the most credible source
Barclays CEO on Bonds, AI and UK Bank TaxBarclays CEO CS Venkatakrishnan discussed pressure in the bond market, how the bank is using AI, and urged the UK government not to raise taxes on banks in the next budget.
xAI Settles Claims Against Apple Over AI CompetitionElon Musk’s xAI and X Corp. resolved an antitrust lawsuit against Apple that alleged the iPhone maker favored OpenAI’s ChatGPT over other chatbot makers.
Why the AI Sell-Off Is a Buying Opportunity: TenglerNancy Tengler says the recent sell-off in AI stocks is a hiccup rather than a collapse, arguing the AI buildout remains a once-in-a-generation opportunity and outlining where winners may emerge.
Ex-DeepMind Scientist: We Must Slow Down AIFormer Google DeepMind researcher Alex Turner left over safety concerns and argues AI progress must be slowed; he proposes tracking compute as a practical regulatory lever and questions Big Tech’s motives when calling to “slow down.”
Carney on Investment, AI Coordination, and US TradeCanadian leader Mark Carney discussed boosting foreign capital investment, seeking coordination on AI safety, and the state of US trade ahead of the national investment summit in Toronto.
Perplexity Portable Computer Now on Windows with NVIDIA RTXPerplexity launched Portable Computer in its Windows app for NVIDIA GeForce/RTX PRO PCs, running local agents accelerated by NVIDIA GPUs. It ships with a local model (e.g. Qwen 3.8 27B), keeps sensitive data on-device, and can escalate tasks to cloud models with user permission.
TechCrunch Disrupt 2026: When OpenAI ships your roadmapTechCrunch will host an interactive Builders Stage session at Disrupt 2026 examining how advancing foundation models can erode startup value and product roadmaps. The event targets AI builders; registering before Sept 25 saves up to $200.
AI bosses risk clash with Wall Street over safetyA Bloomberg Opinion piece says Anthropic PBC and OpenAI, while urging slower AI progress, must weigh pressures from industry, financial markets, and the Trump administration, risking clashes with Wall Street. Columnist Parmy Olson discusses the implications.
Superhuman acquires notetaker FathomSuperhuman acquired YC-backed meeting notetaker Fathom, which reports over 400,000 monthly active users and more than 1 million recorded meetings. Fathom has raised over $30 million and was valued at $94 million in 2024; Superhuman had tested a notetaker internally but opted to buy an established product to integrate AI into its suite.
RubyGems Open-Source Supply Chain Security (HN)A Hacker News thread shares discussion about RubyGems open-source supply-chain security and related OpenAI topics, focusing on ecosystem safety.
King Charles to host AI executivesReuters reports King Charles will host AI executives amid mounting concerns about the threats posed by artificial intelligence.
Hear how AI can engineer nature’s comebackTechCrunch Disrupt 2026 will feature discussions on how AI, genetics, and computational biology are enabling conservation research and efforts to revive extinct species, raising scientific and ethical questions. The topic has attracted billion‑dollar startups.
Xi proposes open-source AI zone for BRICSXi Jinping proposed creating an open-source AI zone among BRICS countries to strengthen cooperation and joint development in AI, a proposal reported by multiple outlets.
China Rebukes Anthropic CEO’s Call to Curb China AIAnthropic CEO Dario Amodei urged limits on advanced AI chip and equipment sales to China and a global slowdown; China’s Ministry of Foreign Affairs rejected fearmongering and called for cooperation. The article notes US and Chinese leaders will discuss AI governance on Sept. 24.
Adversarial Fashion challenges surveillanceThe report describes adversarial garments that disrupt facial-recognition patterns to challenge surveillance norms; while they don’t fully block AI cameras, they can confuse recognition databases. It mentions Urban Privacy and writer Rina Diane Caballar.
HN: Preprompt Migration and Inference Privacy RisksHacker News covers gotchas migrating preprompts from Opus to self-hosted Ollama and warns that running inference on frontier providers can leak session metadata and be used for training; the thread references the Navier–Stokes episode and distrust of OpenAI and Anthropic.
A Vinyl Bar launches music apps, raises $5.5M pre-seedA Vinyl Bar, founded by former Spotify innovation lead Máuhan M Zonoozy, is releasing a series of small interactive “singles” apps that let users modify and create sounds. The startup closed a $5.5M pre-seed round with investors and advisor Dawn Ostroff. The product strategy focuses on shifting listeners from passive consumption to active music participation.
The AI job market in 2026A Hacker News summary compiles major sources on the 2026 AI labor market: AI Engineer is the most hired role, Research Scientist the most prestigious, with agentic systems and Forward Deployed Engineering growing fastest. It notes large 2025 increases in AI job postings and a substantial wage premium for AI skills.
Report ties Anthropic to Rumble’s $13.7B compute deal, stocks reactA media report named Anthropic as the customer behind Rumble’s reported $13.7B six-year compute contract, sending Rumble shares up ~10% while CoreWeave and Nebius fell ~9%. The arrangement has not been confirmed by the parties.
Elon survey: most Americans fear AI threatens human purposeAn Elon University survey finds 69% of US adults think AI will weaken human-to-human bonds and 64% fear AI replacing tasks will undermine life’s meaning; many daily users report increased dependence and reduced critical thinking.
Pelican-Bicycle Benchmark: 2026 UpdateInspired by Simon Willison, the author reran an SVG-generation benchmark in 2026 using stronger models: 10 of 30 prompts were run through six models on OpenRouter, costing about $20 before stopping.
Trump says he uses AI ‘for a lot of things’ without detailsPresident Trump said he uses AI “for a lot of things” but did not specify how, claiming U.S. leadership in AI will bring benefits. The report also notes Anthropic CEO Dario Amodei's warnings about AI agent risks and concerns voiced by former Anthropic researchers.
Fed meets amid uncertainty over oil, AI and inflationThe Federal Reserve met against a backdrop of uncertainty from oil prices, AI developments and inflation, weighing potential impacts on macroeconomic policy. The meeting highlights policymakers balancing new technology risks with traditional economic factors.
UN Urges Urgent Action on AI RisksThe UN human rights office reports that Türk called for urgent, meaningful action to address unprecedented AI risks, stressing the need for swift global responses to emerging AI threats.
Clay Institute says Navier–Stokes problem 'apparently settled'The Clay Mathematics Institute said the Navier–Stokes Millennium Prize Problem has "apparently been settled" and a formal review is now underway. The announcement indicates the claim has moved into verification but is not yet finalized.
Microsoft publishes 'humanist' AI code of conductMicrosoft published a 37-page humanist AI code of conduct that says "people matter more than AI," asserts models are not conscious and should not imitate consciousness, and rejects pursuing legal personhood for AI. The move responds to growing safety concerns and recent calls by figures like Anthropic's Dario Amodei to slow AI development.
llama.cpp adds Maple 20B-A1B ternary MoE (CPU)A PR to ggml-org/llama.cpp adds the Maple 20B-A1B ternary MoE architecture by AlexGabbia, runnable on CPU and visible as deepgrove/maple-preview on Hugging Face. The change extends open-source options for users with low VRAM.
Code shows Apple may let Siri use third‑party modelsLeaked code for iOS 27/macOS suggests Apple's new Siri architecture supports third‑party model delegation and replacing server‑side inference, with examples showing Claude and ChatGPT as Siri extensions and a potential server swap to GPT‑5.6. If real, this could enable deep integration of external large models with Apple system features.
Kinesis — control your Mac with the Meta Neural BandA developer repurposed the Meta Neural Band's sEMG data in an independent project called Kinesis to control macOS via gestures—switching desktops, opening Mission Control, and adjusting volume. The project is experimental, tested only by the author, and intended to encourage further exploration.
The A.I. Slowdown Debate Goes GlobalThe New York Times ran a piece titled “The A.I. Slowdown Debate Goes Global,” discussing international debate over slowing AI development; no further details provided in the excerpt.
The A.I. Daily