Artificial Intelligence News

Google Research senior researcher Alex Krentsel independently reproduced Experiment 5 from MIT CSAIL's weight-space programming blog post and found the ability to hand-write a working neural network's weights appears onl…

A developer outlined a four-layer "Physical AI" architecture for closing the edge MLOps gap in manufacturing and logistics, combining an identification engine (RFID, ultra-wideband RTLS, BLE tags), an edge sensing engine…

Every one of twelve frontier models from nine labs billed more for the same content in Swedish, Norwegian, Danish and Finnish than in English, according to a measurement of 3,000 sentences from the FLORES-200 and NTREX-1…

Butterneck, a tool that lets product managers and other stakeholders request code changes through AI coding agents like Cursor, Claude Code and Antigravity, deliberately stops short of opening pull requests, merging or d…

Mav AI has launched a match-analysis product that converts full soccer match footage into tactical reports, key clips and next-game recommendations in under 30 minutes, targeting semi-professional and lower-division club…

A paper coauthored by Alex Zhang demonstrates "Weight Space Programming," in which a large language model directly reads and edits the weights of a smaller model, including writing entire network weights from scratch in …

A 2021 paper by Elhage et al. took initial steps toward reverse-engineering transformer language models, finding that "induction heads" — attention heads that only develop in models with at least two attention layers — c…

Vercel released agent-desktop, a CLI and C library that reads the accessibility tree of a desktop app and returns it as JSON so AI agents can act on UI elements by ref, shipping for macOS 13 or later, Windows 10 1809 or …

DeepSeek's Harness team showcased Creator Mode on November 2, letting users build custom plugins for the open-source agent platform through natural-language conversation. DeepSeek published seven demonstration prompts in…

ActivTrak research published in 2026 found that the heaviest AI users among 9,140 knowledge workers at 467 enterprises worked between 8.42 and 9.08 hours per day over April 1 to September 25, 2026, versus 7.58 hours for …

Re-running an agentic coding evaluation with GPT-6 Astra, the author found that agents achieved full holdout test correctness on a Zstd-from-RFC implementation over 99% of the time at medium effort for $1, versus about 6…

A developer built a retrieval-augmented generation service with hybrid dense+BM25 retrieval, a 0.15 relevance threshold meant to trigger refusal, and an extractive mock generator that quotes cited chunks, then evaluated …

IBM published a security bulletin on October 28, updated October 29, disclosing four vulnerabilities in Langflow, the open-source tool for building AI agents and workflows, including two critical flaws scored 9.8 that re…

Google will double Gemini 4 Argon API prices on December 1, raising input tokens from $2 to $4 per million and output, including thinking tokens, from $10 to $20 per million, per the model's official documentation. Argon…

Mistral released the open weights of Mistral Large 4, a roughly 1-trillion-parameter natively multimodal Mixture-of-Experts model with a 1M-token context, on Hugging Face on October 29, 2026. The release ships under the …

Meta's Muse personal AI agent has paid roughly $1.2 million in bug bounties, including $300,000 for an authentication flaw found four months before launch, David Singleton said in a thread on X and an accompanying blog p…

Anthropic's latest S-1/A filing, published Monday, adds a Risk Factors warning that the company "may imminently meet our Responsible Scaling Policy's definition of 'dramatically accelerated progress attributable to AI,'"…

The UK National Cyber Security Centre (NCSC), part of GCHQ, published its State of the Cyber Nation report on 3 November 2026 assessing it is "highly likely" that AI-powered malware will be used in a ransomware attack on…

Businesses using AI in the EU face only limited obligations today under the EU AI Act — staff AI literacy, avoiding banned practices, and labeling two kinds of content — while the heavier deployer duties for AI used in h…

A Stanford University study published November 2, 2026 found that 44% of 50 widely cited LLM findings no longer hold when retested on 20 models released from July 2024 to September 2026 across eight developers, with anot…