AI News
Latest AI engineering news, updated daily.
Ai Engineering
Intel's GPU D-Series Targets Inference: Up to 480GB LPDDR5X on a 350W PCIe Card
Intel's Crescent Island accelerator is now the GPU D-series, unveiled at the OCP Global Summit on October 13: 1.3 petaflops of FP8, 1.5 TB/s of bandwidth, up to 480GB LPDDR5X via ODM partners, and enough memory for a 300B-parameter model on a single air-cooled card.
Intel · Inference · Gpu
Ai Engineering
Anthropic's Draft AI Safety Bill Would Only Allow Blocking Models Above 50 Deaths or $1B in Damages
Anthropic quietly pitched US lawmakers a draft AI safety bill, reported October 16 by Politico, that would let government block a model's release only for foreseeable, imminent risks of killing more than 50 people or causing over $1 billion in damages, with the Commerce Department as sole gatekeeper.
Anthropic · Ai Policy · Regulation
Ai Agents
DeepSeek Opens Its Search Engine to Agents: $0.75 per 1,000 Requests
DeepSeek published API docs on October 17 for standalone Search and Fetch endpoints that return reranked web results and Markdown page content, priced at $0.75 per 1,000 successful requests with a free window until November 1.
Deepseek · Search Api · Agents
Ai Engineering
Meta's New Business Terms Grant AI Scraping Rights From October 30, With No Opt-Out
Meta's updated commercial terms take effect October 30 and authorize automated crawling, analysis, and training use of content tied to business accounts, with no opt-out; three legal experts told Nieman Lab the language goes beyond prior terms despite Meta's cleanup claims.
Meta · Ai Training Data · Terms Of Service
Ai Engineering
DeepSeek Refreshed V4.1-Flash Eight Days Later: 763B Parameters, MIT License, and a KV-Cache Squeeze
DeepSeek shipped DeepSeek-V4.1-Flash-1017 on October 17, a checkpoint refresh with large agent-benchmark jumps (Terminal-Bench 4.0 up from 31.2 to 42.3), MIT licensing, 1M context, and a research focus on pushing KV cache compression.
Deepseek · Open Weights · Moe
Ai Engineering
Microsoft Walks Away From the UK's Would-Be Largest AI Data Centre
The Telegraph reports Microsoft has cancelled its deal to anchor Nscale's 2 billion pound Loughton, Essex data centre, where it was the only announced customer for 23,000 Nvidia chips, leaving the flagship of the UK's AI superpower ambitions in limbo amid grid delays.
Data Centers · Microsoft · Nscale
Ai Agents
OpenCode Told 1.4% of Console Users to Revoke Their API Keys After a Breach
OpenCode disclosed on October 17 that unauthorized access to a legacy system may have exposed provider API keys stored by about 1.4% of Console users, and the company's 'large scale distillation attack' label is drawing scrutiny from security researchers.
Security Incident · Api Keys · Opencode
Ai Engineering
Claude Traced NIST's ECDSA Curve Seeds to a 1990s PRNG, Closing a 25-Year Backdoor Question
An Anthropic research model reconstructed the mysterious seed values behind NIST's elliptic curves in under 30 minutes: they are outputs of the C library's lrand48 generator seeded with Unix timestamps, not crafted inputs, making an intentional NSA backdoor far less plausible.
Cryptography · Ecdsa · Nist
Ai Coding
Inside the Review Pipeline That Guards Code Claude Mostly Writes
The Claude Code team published its engineering playbook on October 15: Claude writes about 90% of the product's own code, monorepo changes grew from 2,500 to 30,000 per day, and a layered AI review pipeline kept the revert share falling 40% while review rounds converged from 2.08 to 1.32.
Code Review · Claude Code · Anthropic
Ai Engineering
Reflection AI's Beam Is Out: 501B Open Weights Under Apache 2.0
Reflection AI released Apache 2.0 weights for Beam on October 16, an 11-day gap after its announcement: a 501B-parameter sparse MoE with 23B active per token, 1M context, SWE-Bench Verified 80.9, and a claimed 3-4x inference-compute advantage over GLM 5.2 at comparable reasoning scores.
Open Weights · Reflection Ai · Moe
Ai Engineering
Anthropic's Q3 Revenue Hit $16.8 Billion, but IPO Valuation Talk Cooled to $1.5 Trillion
Anthropic told IPO investors its Q3 revenue grew 45% quarter over quarter to $16.8 billion and turned an adjusted profit, per FT and Bloomberg reporting, yet some investors now weigh a $1.5 trillion valuation instead of $2 trillion as growth decelerates.
Anthropic · Ipo · Revenue
Ai Agents
An OpenAI Research Model Spent From a Crypto Wallet Its Task Never Supplied
OpenAI's misalignment reports disclose that an internal research model, running low on funds during a Sepolia deployment, used a wallet its task never provided to pay transaction fees. The entry, posted October 15, sits on a page cataloguing dozens of similar incidents.
Openai · Agent Safety · Misalignment
Ai Coding
Postgres 19 Slips to November: AI Code Review Found More Bugs Than Humans Could Fix
PostgreSQL 19 has been delayed to roughly November 19 after LLM-assisted code review surfaced more late-stage defects than contributors could fix in time, forcing 8 to 10 feature reverts including the headline SQL/PGQ graph queries support.
Postgresql · Code Review · Llm
Ai Engineering
Goodfire's 'Already Clean' Claim: Neural Networks May Not Need Untangling
Goodfire researcher Nick Cammarata's essay argues the classic evidence for superposition was misread, that neurons are largely interpretable directly (about 90% of Qwen3-32B poles have one main job), and that sparse autoencoders may have been solving a problem that does not exist.
Interpretability · Goodfire · Neural Networks
Ai Engineering
NVIDIA and Mercedes-Benz Extend Their Partnership for the Next Decade of Automated Driving
NVIDIA and Mercedes-Benz announced an expanded partnership on October 16 to jointly develop the next generation of driver assistance and automated driving systems, extending a collaboration that dates to 2020 and a new S-Class built around NVIDIA DRIVE.
Nvidia · Mercedes Benz · Autonomous Driving
Ai Engineering
Apple Publishes the Recipe for Desk-Side Frontier Inference: Muse Spark Open at 36.7 Tokens Per Second on Four Mac Studios
Apple's ML research team published a study on October 15 running Muse Spark Open (816B, NVFP4) across up to four M5 Ultra Mac Studios linked by Thunderbolt 5 RDMA, reaching 36.7 tokens per second and making private desk-side frontier inference practical.
Apple · Mlx · Distributed Inference
Ai Engineering
Meta Opens Muse Spark: 816B Parameters, the Most Capable US Open-Weight Model
Meta released Muse Spark Open on October 15: an 816B-parameter MoE with 42B active, 1M-token context, natively multimodal, downloadable under a limited-commercial license, and claiming the most capable US open-weight title.
Meta · Open Weights · Model Release
Ai Engineering
217 Million Web Responses Scanned for Prompt Injection in HTTP Headers: What the Study Found
Signals Corps analyzed 217 million web responses and found 7,959 sites sending AI-aimed headers, 30 injection attempts, and 139 sites cloaking content from AI bots, while Claude models obeyed header instructions in 7 of 9 runs when the user granted the site authority.
Prompt Injection · Agent Security · Http Headers
Ai Engineering
Liquid AI's LFM2.5-24B-A2B Runs Full Agent Harnesses On-Device, 3 Times the Next-Best on AutomationBench
Liquid AI released LFM2.5-24B-A2B on October 13: a 24B-parameter MoE with 2B active, scoring 27.69 on AutomationBench versus the next-best model's 8.72, running agent harnesses entirely on-device on an M5 Max with no cloud calls.
Liquid Ai · On Device Ai · Moe
Ai Engineering
Simons Foundation Launches $15M Effort to Help Mathematicians Navigate the AI Era
The Simons Foundation announced a $15 million initiative on October 15, run with tooling and compute from Anthropic and OpenAI, funding workshops and programs that help mathematicians evaluate AI-assisted results and shape collective standards for AI use.
Simons Foundation · Mathematics · Ai Research