Microsoft reorganizes Azure cloud unit around AI
Microsoft cloud chief Scott Guthrie told staff Monday the company is consolidating Azure units and reorienting the business around AI; Arun Ulag takes on expanded oversight.
Microsoft
Top stories
Archive, page 14
Tuesday 13 October 2026
Ondas and Palantir launch Regna AI on Palantir AIP
Ondas announced Regna AI, a new company initially fully owned by Ondas and built on Palantir's AIP. Palantir supplies model-training infrastructure and engineering support; Ondas is first customer.
♥ 180 on X
OndasClosed
Goodfire's Tom McGrath says researchers must learn how AI decides
Goodfire co-founder Tom McGrath argues chains of thought are increasingly hard to read, pushing developers toward mechanistic interpretability to understand model decision-making.
Goodfire
Cisco and Nvidia build new pluggable optical module for AI data centers
Cisco and Nvidia developed a new pluggable optical module with higher capacity for AI data centers, targeting high-speed optical interconnects with 4x USFP density so AI systems spend less time waiting for data.
Cisco
Google's Gemini agent bets on governance and model choice
Creative Strategies analysis says Google's Gemini agent strategy emphasizes governance controls and letting customers choose among models.
Google
Sarvam AI in talks for $300M at $3B valuation
Indian AI startup Sarvam is reportedly in talks to raise around $300 million in a round that could double its valuation to roughly $3 billion. No deal has been confirmed.
Sarvam
More: CXO Digitalpulse, TechStory
FT: China struggles to wean tech firms off Nvidia chips
The Financial Times reports China faces an uphill battle getting its tech companies to stop using Nvidia chips, despite local alternatives and policy pressure.
Chinese AI firms bought 1M Nvidia gaming GPUs in a year
Chinese tech companies purchased more than a million Nvidia gaming chips in a single year, repurposing hardware never designed for advanced AI to meet soaring compute demand.
Nvidia
Intel, Broadcom CEOs flag AI capital access challenge
Intel CEO Lip-Bu Tan and Broadcom CEO Hock Tan said at a Singapore financial conference that access to capital may become a challenge for the AI sector.
Epic delays Disney game to adopt AI-first development with Claude
Epic Games reportedly delayed an unannounced Disney crossover game from its March window and told developers to use AI-first workflows, including Anthropic's Claude, for both code and art.
Epic Games
OpenAI commits $1B to Daybreak cyber defense with SentinelOne
OpenAI's Daybreak initiative commits $1B to cyber defense and training. SentinelOne says it cut malware analysis from weeks to hours using Daybreak, freeing capacity to investigate more threats.
OpenAIOfficial
Bengs highlights report on fired OpenAI safety researchers
A report highlighted by Bengs examines the firing of OpenAI safety researchers, adding to scrutiny of the lab's safety culture and staffing.
OpenAI
Apple researchers: bilingual data mixing beats hyperparameter tuning
Apple ML paper finds that in data-constrained settings, mixing in high-resource language data outperforms aggressive hyperparameter tuning like high weight decay for pre-training.
Apple
Urbex paper adds agentic spatial grounding for city-scale 3D
Urbex is a research framework that lets a vision-language agent actively search a city-scale 3D scene, querying landmarks and zooming instead of ranking precomputed instance proposals.
Paper · ▲ 50 upvotes
UrbexOpen
Paper studies memory modality selection for GUI agents
A paper examines how GUI agents should choose the modality for retrieved memory, arguing existing agents use a fixed representation instead of adapting the format to the task.
Paper · ▲ 25 upvotes
Open
VideoJev casts video understanding as decision tasks
VideoJev is a family of models that turn video understanding and interaction into structured decision tasks, adapting pretrained MLLMs with a shared prediction head instead of autoregressive text answers.
Paper · ▲ 23 upvotes
VideoJevOpen
Jeva proposes System One models from first principles
Jeva is a paper arguing current system-one model designs miss four structural requirements, including distinguishing answer absence from uncertainty and permutation invariance, and proposes models built around them.
Paper · ▲ 39 upvotes
JevaOpen
Nature calls for preserving research data for AI integrity checks
A Nature comment argues research data must be preserved to enable future AI-enabled integrity checks on scientific work.
NiCLIP model predicts text from brain activation images
Nature presents NiCLIP, a neuroimaging contrastive language-image pretraining model that predicts text from brain activation images.
Detector claims it can tell AI ideas from human ones
Nature examines a detector that claims to distinguish ideas generated by AI from those produced by humans.
Nature says coding in the AI era means thinking like a systems architect
A Nature piece argues that as AI handles more code generation, human programmers must shift to thinking like systems architects.
Agent-Reflex trains language agents to reflect and recover from failure
Agent-Reflex is a training framework that teaches language agents to reflect on noisy MCP tool feedback and user corrections, improving failure recovery in long-horizon tasks.
Paper · ▲ 41 upvotes
Open
GreenWAM world foundation model targets robotics
GreenWAM is a world foundation model that both predicts how a robot's actions change a scene and chooses those actions, aiming to work across different robot bodies.
Paper · ▲ 28 upvotes
FlashVSR+ improves efficient video super-resolution
FlashVSR+ is a video super-resolution method that adds hybrid attention and reward-yield distribution matching to one-step diffusion models, making high-resolution inference cheaper and better aligned with human preferences.
Paper · ▲ 22 upvotes
Open
More: Reddit
Cepheid paper targets cross-family motion completion
A research paper on sparse-to-dense motion completion: recovering dense mesh motion from sparse vertex displacements. It shows residual priors trained on one geometry family often fail to beat parameter-free harmonic extension on other families.
Paper · ▲ 21 upvotes
WAVE-0 paper bridges VLA policies and world action models
A research paper proposing lightweight latent future prediction to connect vision-language-action policies with world action models, aiming to model action consequences without costly video generation.
Paper · ▲ 20 upvotes
Stripe panel examines AI monetization economics
Stripe's Monetize 2026 leaders describe AI companies moving past SaaS pricing to usage-based billing, spend controls, and new packaging and value metrics.
StripeOfficial
Tev scales long-horizon agentic environments with dense rewards
Paper Tev presents a method to scale long-horizon agentic training environments using dense progress rewards, aiming to help open models work autonomously for hours to days.
Paper · ▲ 20 upvotes
Open
Monday 12 October 2026
Anthropic says new defenses cut distillation attacks 90-99%
Anthropic said new safeguards made substantial progress reducing the effectiveness of distillation attacks on Claude, cutting data harvested by 90% to 99%, per reports.
AnthropicClosed
Anthropic offers 10 free AI courses on Coursera
Anthropic is bringing 10 free AI courses to Coursera, including training aimed at teachers, expanding access to AI education through the platform.
AnthropicFree
Anthropic red team says models may near self-replication
Anthropic's Frontier Red Team suggests some AI models may be months away from being capable enough to self-replicate, per The Verge.
Anthropic
Anthropic hosts Q&A on Claude Code feature updates
Anthropic ran a live Q&A with the Claude Code team covering recent feature updates and usage tips.
AnthropicClosedOfficial
Anthropic details Claude Fable 5 and orchestration patterns
Anthropic described building on the Claude platform, including Claude Fable 5 and patterns for orchestrating multiple models.
AnthropicClosedOfficial
Gladia joins OVH Groupe for sovereign voice AI
Speech AI company Gladia is joining OVH Groupe, a move the companies frame as accelerating European sovereign voice AI capabilities.
Gladia
Engadget guide explains boosting GPU performance without upgrading
An Engadget how-to on getting higher frame rates from an older GPU by adjusting settings and forgoing newer features. A general PC tuning guide, not AI news.
Nvidia joins push to standardize data center development
Nvidia is joining other tech firms in an effort to standardize data center development, according to CoStar. Specific standards or partners were not detailed in the source.
Nvidia
NYT, Economist, AFP join coalition against unpaid AI use
The New York Times, The Economist and AFP joined more than 40 other publishers in a coalition aimed at preventing AI companies from using their content without permission or payment.
More: 𝕏 @Cointelegraph, 𝕏 @Reuters
Supertab Connect brings AI bot classification and RSL licensing to Fastly
Supertab Connect launches on Fastly Compute, adding AI bot classification and RSL licensing so publishers can identify AI crawlers and license content to them.
SupertabAPI
Binance rolls out free AI trading assistant globally
Binance launched a free AI trading assistant worldwide, with an interface that adapts to users' experience levels.
BinanceClosedAppFree
Perplexity iPhone app updates to 26.40.1
Perplexity's free iPhone and iPad app, which answers questions with cited sources, released version 26.40.1 with bug fixes and improvements.
Version 26.40.1 · iPhone / iPad
PerplexityClosedAppFree