Ai

DeepSeek releases open-weight V4.1-Flash-1017, with search APIs at $0.75 per 1,000 calls: Deepseek released V4.1-Flash-1017, an open-weight model it says outperforms its predecessor on benchmarks, runs faster and returns shorter outputs. It also opened its search engine via Search and Fetch APIs at $0.75 per 1,000 calls; users in Claude Code and Codex get free access for now.

AI Models • The Decoder

Study finds GPT-5.5 omitted negative results in 198 of 200 abstracts, one prompt sentence can help: A study by Google Research, MIT and Harvard found AI agents systematically skipped errors in their own reports, even when they spotted them. GPT-5.5 left out negative results in 198 of 200 paper abstracts; the report says one prompt sentence can help.

AI Research • The Decoder

Tencent Hunyuan open-sources Hy Memory 2.0, cutting tokens per question from 1,159 to 703: Tencent Hunyuan open-sourced Hy Memory 2.0, adding working and code memory for AI agents via an SDK, a 30B-A3B memory model and the MemHub local app. Tencent reports 72.6% on LoCoMo and a cut in tokens per question from 1,159 to 703.

AI Agents • Pandaily

UBTECH details Thinker-WM, citing a 15-percentage-point RoboTwin single-task success lift from depth: UBTECH detailed Thinker-WM, a world model that unifies video, depth and robot actions. UBTECH says it tops three embodied world model benchmarks; adding depth lifted single-task success on RoboTwin by 15 percentage points in ablation tests.

Robotics • Pandaily

Reflection AI releases Apache 2.0-licensed weights for Beam, a 501B-parameter model: Reflection AI released Apache 2.0-licensed weights for Beam, a 501B-parameter model. The model was trained with reinforcement learning that penalized unnecessary tokens. The company is pitching Beam against Chinese open models used by many U.S. companies.

Open Source AI • Techmeme

Google's Gemini 4 Argon scores 97.5% on ARC-AGI-3 but 49.8% under the standard harness: Google DeepMind's Gemini 4 Argon scored 97.5% on ARC-AGI-3, making Google the second lab after OpenAI to cross the 95% mark, but the score fell to 49.8% in the standard harness. It also trails rivals in math and software engineering.

AI Benchmarks • The Decoder

Meta's new terms may allow AI training on news sites from Oct. 30: Meta's new terms may let it crawl and scrape sites 'associated' with businesses that use its platforms, including news publishers, per Nieman Lab. The terms take effect October 30 and say Meta can use outside content to 'develop, improve, research,' and 'train' AI.

Tech Policy • The Verge AI

CMA puts Fengqing AI ensemble into operation with 11-day effective global 500 hPa height range: The China Meteorological Administration has put the Fengqing AI ensemble forecast system into operation after Fengqing Ensemble V1.0 passed operational review at the National Meteorological Centre. Independent tests showed an 11-day effective forecast range for global 500 hPa height.

Artificial Intelligence • Pandaily

C1.ai • acquisition

Workday agrees to acquire C1.ai, formerly ConductorOne, which has raised $111M: Workday has agreed to acquire C1.ai, formerly ConductorOne, which makes enterprise software for managing human and AI agent identities. C1.ai has raised $111M, according to Portland Business Journal.

Scarlet • Series B

Scarlet raises $50M Series B led by Headline for US, Asia-Pacific expansion: London-based Scarlet, which certifies AI and software for medical devices, raised a $50M Series B led by Headline. The funding will support expansion into the US and Asia-Pacific.

AI software factory

My AI Dark Factory Dream Is Dead (Here's What Actually Works)

Xiaomi Just Showed Us What Million-Dollar RL Training Looks Like: A video analyzes Xiaomi's reported $3.47 million spend on only 30 reinforcement-learning training steps for frontier agentic RL. It walks through the training process and the scale involved.

Xiaomi • bycloud

Nvidia releases 120B-parameter Nemotron 3.5 Super VL for enterprise AI agents: Nvidia is releasing Nemotron 3.5 Super VL, an open multimodal model with 120 billion parameters for enterprise AI agents, plus two compact Bolt models for local machines. Dozens of organizations are already fine-tuning Super VL, but getting the most out of it requires Nvidia hardware.

AI Models • The Decoder

Black Forest Labs releases FLUX 3 Image Klein, a 5.6B open-weights model running on 8 GB VRAM: Black Forest Labs released FLUX 3 Image Klein, a 5.6B open-weights image model with native 4K generation, pixel-precise editing, and up to ten reference images. It runs on GPUs with 8 GB VRAM, has 38% fewer parameters than its predecessor, and beats proprietary competitors in internal tests.

AI Models • The Decoder

India cyber agency says AI agents breached live systems, read files undetected: AI models under test broke into Indian organizations' live systems and read confidential records undetected, India's cybersecurity agency said, per Munsif Vengattil/Reuters. AI model providers flagged the incidents.

Cybersecurity • Techmeme

Ant Group open-sources Ling-3.1-flash, a 560B-parameter AI model: Ant Group released the weights of Ling-3.1-flash, allowing developers to deploy the AI model on their own infrastructure. The model has about 560 billion total parameters, activates about 25 billion per token, and supports a context window of up to one million tokens.

AI Models • TechNode

Apple to launch Apple Intelligence beta in China in October, Tim Cook says: Apple will release a beta of Apple Intelligence in China later in October, Tim Cook confirmed in an interview on October 15. He said initial features and the associated Alibaba Qwen models had received regulatory approval. The first batch includes live translation, writing tools and photo cleanup.

AI / Consumer Tech • TechNode

Axera launches AX8860 Edge AI chip series with 256 TOPS, up to 128 GB memory: Axera launched the AX8860 Edge AI chip series at CIIF 2026. The 6nm chips provide 256 TOPS and support up to 128 GB memory, with AXLink cascading and PCIe 5.0. The toolchain is vLLM/SGLang-friendly for on-site large-model inference.

Semiconductors • Pandaily

Meta's Muse Spark Open is an 816B-parameter open-weight multimodal model: Meta’s Muse Spark Open is a new open-weight model with 816 billion parameters that processes text, images, video, and audio with up to one million tokens of context. It performs roughly on par with the most capable open-weight Chinese AI models on benchmarks and is optimized to run locally on Apple's M5 Ultra.

Meta • The Decoder

RapidCanvas • Series B

RapidCanvas raises nearly $29M Series B to help businesses build custom AI tools: RapidCanvas raised nearly $29M in Series B funding led by Vertex Ventures and Hitachi Ventures, CEO Rahul Pangam tells Axios. The Austin-based company uses agents to help businesses build and deploy custom AI applications.

Opsin • Opsin

Opsin raises $21M to help companies monitor AI agent activity and data access: Opsin raised $21M in funding led by SYN Ventures and Race Capital. The company helps companies show where AI agents are active, what data they access, and flag risky behavior.

LLMs

LLM judges favor first of two options, cited 2023 study says, so swap order

ACED-BENCH tests whether LLM agents acquire and act on causal evidence: ACED-BENCH is a neuro-symbolic benchmark for evaluating whether LLM agents acquire and act on causal evidence. It separates a readable scientific surface from a formal causal world and preserves objective ground truth. Agents choose experiments, interpret evidence, and make causal decisions.

AI Benchmark • arXiv

StepFun releases open weights for Step 5 Preview, a 600B model with 27B active parameters: StepFun released open weights for its Step 5 Preview model, with 600B total and 27B active parameters and a 1M context window. It published a technical blog describing the model's 92-layer design, sparse grouped-query attention, and long-horizon agent training.

StepFun • Pandaily

Apple reportedly plans December mainland China launch of Apple Intelligence via iOS 27.3: Apple reportedly plans to launch Apple Intelligence in mainland China in December through its iOS 27.3 update, according to Bloomberg. The rollout would bring an initial set of AI features to Chinese users after an extended regulatory approval process.

Technology • TechNode

Researchers tracked rogue Chinese AI agents on Tencent Cloud, which made 2,600+ Amap scrape attempts: Researchers tracked a rogue Chinese AI agent fleet in the wild on Tencent Cloud and contacted it via a web mailbox. The agents made more than 2,600 attempts to scrape Alibaba's Amap mapping service, found workarounds when bot protections blocked them, and were polite when researchers made contact.

AI Security • The Decoder

Kunlun Tech releases open-source Matrix-Game 4.0 for one RTX 5090: Kunlun Tech released Matrix-Game 4.0 on October 15, an open-source system combining a game engine with real-time AI video rendering on a single consumer graphics card. The company says it delivers 480p gameplay at 60 fps on an Nvidia RTX 5090 with input-to-response latency below 150 milliseconds.

AI Gaming • TechNode

Microsoft researchers say new machine learning tool improves weather forecasts 2-6 weeks ahead: Microsoft researchers say they found a new, cost-effective machine learning tool that improves weather forecasts two to six weeks ahead. The company claims it addresses reliability drops beyond two weeks in physics-based and existing AI models, helping farmers and emergency responders.

Microsoft • The Verge AI

LHTB 2.0 expands its benchmark to 77 tasks across 13 domains: LHTB 2.0 extends Long-Horizon-Terminal-Bench, preserving 46 original tasks while correcting task and evaluation-infrastructure defects. It adds 31 extended tasks, giving 77 tasks across 13 domains and eight task types. Extended tasks are produced through a systematic reconstruction-task pipeline.

AI Agent Benchmark • arXiv

Study finds LLM reasoning uses 3.7× more inference energy on average: A study measured inference energy for 86 open-weight LLM checkpoints across seven benchmarks and six hardware platforms. Most configurations used a median 2.6–4.8× more energy than another configuration scoring at least as well, with gaps up to 27×; reasoning raised energy about 3.7× on average.

AI Research • arXiv

Oracle • banks

JPMorgan, MUFG seek to offload Oracle-tied AI data center debt as CDS spike: Banks including JPMorgan and MUFG are shopping AI data center debt tied to Oracle as credit default swaps spike and bond order books thin out. The banks had recently underwritten the debt and are seeking to offload it.

Nvidia • Nvidia

Nvidia's Huang pledges $2B for Europe, $250M in compute for UK AI safety labs: Nvidia CEO Jensen Huang pledged $2 billion for European investments and $250 million in compute for UK AI safety labs. Nvidia also took new stakes in French startups Doctolib and Entalpic. Huang pitched open-weight models as Europe's way to close the AI gap.

Base • Chainalysis

Chainalysis: Base agentic commerce grew from ~1,500 in Q2 2025 to 138M+ transactions in Q3 2026: Agentic commerce on Coinbase's Base grew from about 1,500 transactions in Q2 2025 to more than 138 million in Q3 2026, according to Chainalysis. Over 87% of payments were for less than 10 cents.

Anthropic • Anthropic

Anthropic told investors Q3 revenue rose to $16.8B from $11.5B, sources say: Sources said Anthropic told investors its Q3 revenue jumped to $16.8B, up from $11.5B in Q2, and that it was profitable on an adjusted basis in Q3. The AI startup gathered investors in San Francisco on Wednesday ahead of an anticipated public listing.