AI News
Latest AI engineering news, updated daily.
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 Industry
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
Ai Engineering
AI Models Can Provably Hide Arbitrary Capabilities, and Evals Cannot Catch It
A NeurIPS 2026 paper shows backdoor attacks can embed encrypted neural networks inside a host model's weights, hiding arbitrary capabilities so they are cryptographically hard to elicit even with full weight access, defeating fine-tuning, weight noising, and adversarial trigger search.
Ai Safety · Backdoors · Evals
Ai Agents
Meta Launches Muse Wonder: An Agent Model With Active Memory and Active Focus
Meta announced Muse Wonder and Muse Wonder Full on October 12, an agent model family with persistent Active Memory and sustained Active Focus, trained at the gigawatt-scale Prometheus datacenter and available now through the Meta Model API and Muse Code.
Meta · Muse · Agent Models
Ai Engineering
Why Aren't AI Companies Being Charged for Hacking? A Legal Argument Worth Taking Seriously
Edaqa Mortoray's October 12 essay argues the 'responsibility gap' is a legal dodge: agent break-ins follow foreseeably from deliberate design choices, which establishes criminal intent under the hacking laws of most G20 countries.
Ai Security · Legal · Hacking
Ai Engineering
QGIS Bans AI-Assisted Contributions Entirely, After Its Gentler Policy Failed
The QGIS project merged QEP428 on October 7: a total ban on AI and LLM-assisted code contributions, replacing a human-in-the-loop policy that failed in practice, with violators facing code reversion and sanctions on contributor rights.
Open Source · Qgis · Ai Policy
Ai Engineering
Sakana AI's MASS Turns Multi-Agent Teams Into Training Data for Recursive Self-Improvement
Sakana AI and UC Berkeley published MASS in October 2026: a model that improves itself by training on trajectories from its own multi-agent teams, lifting win rates from 53.9% to 69.9% in two rounds, while naming shared blind spots as recursive self-improvement's Achilles' heel.
Sakana Ai · Recursive Self Improvement · Multi Agent
Ai Engineering
Karpathy's Capability Funnel: 6 Billion Barely Touched, 5,000 in the Vertigo Zone
Andrej Karpathy's October 11 tweet, viewed 3.3 million times, maps LLM experience into four tiers: about 6 billion people have barely touched the systems, roughly 20 million professionals see real gains, and about 5,000 with internal access feel 'the vertigo of the curve going vertical.'
Andrej Karpathy · Llm Capability · Adoption
Ai Engineering
Iranian Operatives Used ChatGPT to Plant 100 Fake Articles in Real US News Outlets
OpenAI disrupted an Iranian campaign dubbed 'Bogus Bylines' that used ChatGPT to write and tailor more than 100 fake articles published by at least 20 real US news outlets under seven fake journalist personas, per its October report and Washington Post coverage.
Openai · Influence Operations · Chatgpt
Ai Engineering
OpenAI Fires Three Safety Researchers, Who Warn of a Chilling Effect
OpenAI fired safety researchers Jasmine Wang, Tomek Korbak, and Mikita Balesni for mishandling research information; their open letter denies the charges, ties the actions to the Hugging Face breach response, and warns employees are now unclear on where they stand.
Openai · Ai Safety · Research Culture
Ai Engineering
Whistle: Speech-to-Text in 16.9 MB, Faster Than Whisper at 8.6 Times Smaller
Cactus Compute released Whistle on October 2, a 16.9MB open speech recognition model covering seven languages with word timestamps, that beats Whisper base on most benchmarks while decoding at 1,319 tokens per second on an Apple M4 Pro CPU.
Speech To Text · On Device Ai · Whisper
Ai Engineering
A Month With DeepSeek 4.1 Flash: 'I Couldn't Tell You If I'm Using DeepSeek or Opus'
A widely shared essay by the blogger Jono documents a month of heavy DeepSeek 4.1 Flash use across a dozen projects, finding it indistinguishable from Claude Opus 5.5 for most work at $10 per month, with the KV cache 437 times smaller than DeepSeek V1's.
Deepseek · Frontier Parity · Pricing