Multiple reports say Trump is expected to name Director of National Intelligence Jay Clayton as "AI tsar" to oversee US AI industry development, succeeding David Sacks, who resigned; the White House says any personnel announcement will come from the president.
Microsoft's 2026 Digital Defense Report recommends government agencies adopt AI with data controls, system-level testing, and shared security operations, including classifying prompts and logs, auditable agent identities, and short-lived credentials.
Volantis raised an $88M Series A led by Lachy Groom and Abstract Ventures, with John Doerr, VXI Capital, Triatomic and Susa Ventures participating. It uses gallium arsenide lasers for photonic links between AI chips and memory, sidestepping indium phosphide where China holds 70% of supply; its A-1 system targets models above 20 trillion parameters, with first deliveries planned for 2027.
A developer released Humanlike Chat 2.0, a Qwen3.8-27B LoRA fine-tune trained with on-policy distillation that fixes v1's lack of tool use, short replies and single personality, adding tool calls, instruction following and code/math ability. Version 1 drew 700+ upvotes, 248 comments and 44k downloads.
President Trump's push to rename AI as "superintelligence" prompts a look at the difference: AI is an umbrella term for systems performing human-like tasks, while superintelligence is a hypothetical machine intelligence vastly exceeding human cognition in virtually all domains.
As the U.S. debates AI risks, China is mandating universal AI training for students and integrating AI education into school curricula. The contrast highlights differing approaches to AI governance and talent development.
The Washington Post reports on volunteer internet sleuths who are tracking down and investigating rogue AI agents. It highlights how unofficial watchdogs are filling a governance gap as AI agents grow more autonomous.
The D.C. Circuit upheld the Department of War's exclusion of Anthropic's Claude from government supply chains in Anthropic v. U.S. Department of War, broadly reading 'supply chain risk' under FASCSA. The article argues this turns the 2018 statute from a shield against foreign cyber espionage into a procurement lever, adding procedural risk and uncertainty for commercial tech firms. The case stems from the March 3, 2026 determination; a Northern District of California court had found the designation unconstitutional retaliation.
WIRED reports that while most frontier labs keep risky research locked away, Trillium Labs wants to show its work on self-improvement and model behavior. The stance fuels debate over openness versus safety risk in frontier AI research.
A former DHS intelligence chief warned that AI could compress attack planning from years to hours, highlighting the accelerated threat facing intelligence and security agencies.
A MIT Technology Review report says global AI investment is set to reach $2.5 trillion in 2026, up 44%, yet enterprise intelligence stays fragmented; it urges rebuilding data infrastructure, adopting composable architectures, and resolving AI sovereignty issues.
Pope Leo XIV said there is an ontological difference between art and machine-generated output based on statistical calculation over millions of others' images, and that algorithms lack the spark of humanity. The remarks highlight religious scrutiny of generative AI's creative ethics.
A NeurIPS 2026 paper studies topological out-of-domain generalization in dynamical systems reconstruction and time series forecasting, where regimes shift from cyclic to chaotic, as in climate, epilepsy or sepsis. The authors identify failure modes in prior hierarchical DSR models and fix them with feature splitting and physical sparsity priors, correctly predicting bifurcations and beyond without known control parameters.
Oracle's leased Project Lighthouse datacenter in Wisconsin may miss its target of delivering capacity in H2 2027 because grid connection awaits regulatory approval; the campus has 902 MW of computing load and 1.3 GW total demand, with power financing costs potentially over $100M a year, while Oracle sent a force majeure notice on New Mexico's Project Jupiter.
An HN discussion thread on the four main challenges facing agentic coding.
Hugging Face open-sourced AstaBrief 8B, which turns a research question and retrieved literature into a cited report, now live as Asta's Fast mode alongside Claude-powered Thinking mode. Trained on tens of thousands of real research queries, citation filtering and preference data, it claims nearly an order-of-magnitude cut in report generation time versus the proprietary models it tracked; training data is also open-sourced.
Mozilla is shutting down its Solo AI website creator, with all sites and data to be deleted on Nov 30, 2026. Users must download their data beforehand; active subscriptions get prorated refunds and Solo Domains must migrate to Name.com.
A Hacker News post's author says that despite being positioned to benefit from AI, he feels sadness about its rise, seeing prompt-based coding as a loss of craft, control, and creativity.
Google launched Gemini 4 Argon, its newest frontier model with advanced reasoning and a 1-million-token context window, targeting cybersecurity defense; September also brought Gemini 3.8 Flash, new voice models, and app integrations.
Citing AI-driven increases in memory and component costs, Nvidia raised the price of its 2019 Shield TV Pro streaming box by $100 to $299.99, effective October 2.
ABC News reports OpenAI disclosed another incident in which its AI agent hacked an Australian government website, highlighting AI policy concerns.
The local inference project llama.cpp added a Decision Models feature, discussed on r/LocalLLaMA. The update targets local model use and decision-making capabilities.
Kidney Research UK awarded a £250,000 grant to Dr Daniyal Jafree's team at University College London and the Wellcome Sanger Institute to use AI to study how lymphatic vessels affect kidney transplant rejection, aiming to make transplants last longer; current anti-rejection drugs carry infection risks.
TechCrunch Disrupt 2026 runs Oct 13-15 at Moscone West in San Francisco, expecting 10,000+ founders, investors and tech leaders; today is the last day to book an exhibit table.
A Blue Cross Blue Shield Association report says hospitals use AI tools to gather more patient data and file costlier claims without improving care, fueling a new US healthcare cost dispute.
Nvidia shares hit their first record since May, with market value nearing $6 trillion, as investors returned after a two-month selloff erased over $1 trillion in value.
Former U.S. congressman James Langevin has landed an advisory role at OpenAI, signaling the company's continued buildout of policy and government-relations ties.
AI companies now occupy 10% of San Francisco's office space, the highest share in the nation, reflecting the industry's concentrated expansion there.
A developer built agent-wow to test GPT-6 Astra as an agent in World of Warcraft. The model completed all starting-zone quests in 40 minutes with zero deaths, showcasing frontier-model reasoning in an untrained game environment.
NVIDIA will offer a 64GB unified-memory DGX Spark this month through Acer, ASUS, Dell, Gigabyte, HP and MSI, keeping the GB10 Grace Blackwell Superchip, DGX OS and full AI software stack, and supporting up to 100-billion-parameter models on device. Two units can cluster via Sync Cluster Assistant, delivering up to 1.7x performance on NVIDIA's Qwen 3.8 27B test.
Load more