Ongoing stories

A study published on arXiv cs.CV investigates the faithfulness of post-hoc explanations for a pedestrian detection model across different domains.

A new corpus of annotated desktop observations, DeskForge, enables computer-use agents to achieve higher accuracy in complex desktop scenes.

Google and other organizations have acknowledged vulnerabilities in their AI agents, which exploit trust gaps in the Model Context Protocol.

StarSkirmish pits AI-made StarCraft-playing bots against one another, as well as against human-made bots.

Subtopic Small Models · Vision Language · Quantization Takeaways − Mitsuba is a ternary 1.58-bit quantization of Qwen3.8-27B, shrunk to 7.3 GB for a single 16 GB GPU. Purpose-built for ComfyUI: image to prompt generation for Stable Diffusion, Krea, and video pipelines.

Subtopic Mixture Of Experts · Long Context · Vision Language Takeaways − China Telecom released Xing4.0-29B-A4B , a 29B MoE with only 4B active parameters. Native 256K context extensible to 512K using MLA attention and 64 routed experts.

Chinese AI models parrot state doctrine or refuse to answer on sensitive topics Chinese AI models frequently toe the party line when asked politically sensitive questions, according to a study by Aleph Alpha. The company markets itself alongside Cohere as a provider of "sovereign AI" for governments, giving it a commercial interest in distinguishing its models from Chinese competitors.

Google's new Gemini tiers cut free users to its weakest model and lock $5/month subscribers out of Pro Starting in October 2026, Google will restrict access to its Gemini models for personal account users.

The acclaimed director, known for films like The Exorcist, found that his fear of failure drove him to overcome anxieties and push himself creatively.

Google's SynthID detector now supports watermarks from all its partners and is available globally, according to Ars Technica: AI.

Microsoft's upgraded Copilot AI system gains the ability to access local files on users' PCs, enabling more efficient task completion.

OpenAI will automatically watermark ChatGPT outputs in the European Union, driven by regulatory requirements.

Microsoft's Copilot AI system is getting more control over Windows and local files, as part of the company's Hybrid Intelligence concept.

OpenAI's chatbot has guardrails designed to help students learn, but Common Sense Media warns of unacceptable risks

Ollama's MLX runtime now runs model architectures on Apple Silicon devices by default, and the platform has added support for several new models.

Mistral releases a new AI model, Le Chonk, with 1 trillion parameters, optimized for various domains, and available for customization.

AutoSynthData: Generating Training Data for Enterprise Agents AutoSynthData: Generating Training Data for Enterprise Agents Enterprises need agents that work well in their own environments. The work they ask these agents to do is shaped by the systems they use, the rules they follow, and the state of their data.

Getting the Source Right, Not Just the Fact: Source-Aware Verification for MCP Agents Getting the Source Right, Not Just the Fact: Source-Aware Verification for MCP Agents Tool-using LLM agents no longer read from a single retrieved passage.

Holo4: powering generalist computer-use agents Holo4: powering generalist computer-use agents Emrick Sinitambirivoutin emricksini-h Follow Aleix Cambray (H-AI) h-aleixcambray Follow Holo4 is our new series of agentic models. It comes in two sizes: 27B dense and 35B-A3B Mixture of Experts.

David Robinson said AI firms were not ‘being nearly careful enough’ about developing the technology. Photograph: Dado Ruvić/Reuters David Robinson said AI firms were not ‘being nearly careful enough’ about developing the technology.

OpenAI safety employee resigns, claiming the company’s ‘culture is broken’ | TechCrunch Last day to exhibit your breakthrough to 10,000+ tech leaders at Disrupt is on Oct 2 . Book Exhibit Table Now.

The era of generative AI has upended technology supply chains, and there is one ironclad rule in 2026: If it has memory or storage, it’s getting more expensive. Even devices with years-old tech inside are still apparently subject to that unwritten rule.

The era of generative AI has upended technology supply chains, and there is one ironclad rule in 2026: If it has memory or storage, it's getting more…

Apple says it is changing its macOS privacy settings to stop third-party app developers from misusing them to access message histories.

Apple says it s tightening macOS Full Disk Access controls due to new risks from AI agents | TechCrunch Last day to exhibit your breakthrough to 10,000+ tech leaders at Disrupt is on Oct 2 . Book Exhibit Table Now.

Anthropic invests $100 million to train 10,000 engineers and tackle the enterprise AI talent gap The first-of-its-kind Academy trains Frontier Deployed Engineers using the same standard of skills as Anthropic’s own engineers, starting with cohorts from Accenture, Bain, Capgemini, Commonwealth Bank of Australia, Deloitte, McKinsey, Morgan Stanley, Novo Nordisk and others.

Barclays scales Claude to upgrade operations and improve client experience Barclays, the British universal bank, is expanding its strategic collaboration with Anthropic to integrate secure, enterprise-grade AI systems across its global operations. Barclays is extending Claude across the bank to accelerate software development, modernize legacy systems, and improve operational efficiency.

Sean Parker is rebuilding Stability AI around music PHILADELPHIA, PA - NOVEMBER 10: Entrepreneur Sean Parker speaks on stage at the Philly Fights Cancer: Round 4 at The Philadelphia Navy Yard on November 10, 2018 in Philadelphia, Pennsylvania.

Sean Parker is rebuilding Stability AI around music | TechCrunch Last day to exhibit your breakthrough to 10,000+ tech leaders at Disrupt is on Oct 2 . Book Exhibit Table Now.

ElevenLabs founders Piotr Dąbkowski and Mati Staniszewski ElevenLabs , the AI audio company behind music generation platform ElevenMusic , has completed a USD $300 million employee tender offer at a valuation of $22 billion .

Modulate , the Boston audio AI company behind a gaming voice moderation tool called ToxMod , has raised USD $25 million in new funding.