Researchers from OpenAI, Anthropic, and xAI have published public resignation letters citing internal conflicts between AI safety alignment and commercial pressures to prioritize product development over ethical governance. - Business Insider

OpenAI drops 'safely' from mission statement amid $500 billion for-profit pivot - Business Standard

OpenClaw creator Peter Steinberger is joining OpenAI - X (formerly Twitter)

OpenAI hired Peter Steinberger, creator of OpenClaw, after a bidding war with Meta, to accelerate the development of autonomous personal AI agents that proactively manage workflows and tasks across multiple applications without explicit user prompts. - TNW | Artificial-Intelligence

OpenClaw creator Peter Steinberger joined OpenAI to lead personal agent development following a bidding war with Meta, prioritizing the framework's 196,000 GitHub stars and 2 million weekly visitors over its codebase. - Implicator.ai

OpenClaw demand triggers Apple Mac shortages for high-memory units - Tom's Hardware

Whenever I hear about OpenClaw, this is what comes to mind:

OpenAI, Anthropic, and Cisco joined the Agentic AI Foundation to establish governance standards, addressing infrastructure and trust barriers as Gartner predicts 40% of enterprise software will include AI agents by 2026. - PYMNTS.com

As tasks get more complex, agents break down into a 'hot mess' of cognitive overload. - Anthropic

ByteDance pledged to strengthen safeguards for its AI video tool Seedance 2.0 after Disney issued a cease-and-desist letter regarding the unauthorized use of Marvel and Star Wars characters. - BBC

Paramount sent a cease-and-desist letter to ByteDance, alleging its AI-generated Seedance videos and Seedream images infringed Paramount's IP. Disney also sent a legal demand to Chinese internet company. - Variety

ByteDance is tightening restrictions on its AI video generator following a cease-and-desist letter from Disney regarding a viral deepfake video featuring actors Tom Cruise and Brad Pitt. - Engadget

ByteDance's Seedance 2.0 AI tool produced a hyper-realistic 15-second video of Tom Cruise and Brad Pitt, prompting Disney and the Motion Picture Association to allege copyright infringement and demand the company cease operations. - The New York Times

Larry David + AI ... ngl that's pretty pretty good - X (formerly Twitter)

curb your (artificial) enthusiasm…someone do "alan turing is at his desk dreaming, sees elon musk and grok in his dream, wakes up with a horrified start, eats apple, dies"

6 weeks ago, my socials were all Claude Code, now they are all Seedance

Google places medical disclaimers in AI Overviews behind "Show more" buttons and small fonts, which experts warn prioritizes speed over accuracy and risks the spread of medical misinformation. - the Guardian

Internet videos expose the truth, and cloud it. - The New York Times

Google and Microsoft propose WebMCP standard for structured tool access, like embedding discoverable structured data and artifacts and tools within your web page. - VentureBeat

Google DeepMind paper on the latest AI assisted math research they are doing - arXiv.org

Michael Burry disclosed a short position against Palantir, forecasting a 54% stock decline to the $50-$60 range as the UK's NHS continues investigating the company for allegedly hiring a digital marketing agency to conduct covert smear campaigns. - The Potential Surface

Snowflake CEO Sridhar Ramaswamy warns that software giants risk becoming "dumb data pipes" as AI model makers centralize enterprise data, leading Snowflake to offer flexible access for customers using third-party platforms like ChatGPT. - Business Insider

Polar Capital’s UK fund manager Nick Evans is divesting from application software firms like Adobe and SAP, citing AI as an existential threat that enables clients and startups to build competing internal tools and infrastructure. - Yahoo

do I think there might come a day when all 'software' is plain English instructions to AI and it works perfectly every time? maybe but today is not that day. Until then you need deterministic tools.

Anthropic is launching AI work tools and adaptable plug-ins for professional sectors, including legal and consulting, to compete with established specialized software. - Financial Times

Amazon launched Creative Agent, an AI tool using AWS and Anthropic Claude to automate advertising research and generation, reducing production time from weeks to hours for small businesses in the US and Europe. - TechRadar

An Upwork report shows AI-related skill demand grew 109% year-over-year, led by AI video creation at 329%, while employers prioritize human skills and learning agility for the 2026 market. - Fast Company

Comparative testing of Gemini, Claude, and ChatGPT for natural language app development demonstrated that iterative prompt refinement allows free AI models to match premium performance, making prompt engineering the primary factor in software quality. - CNET

these comparisons are very hard, it's evolving very fast, you need to customize your workflow around the assistant you are using, install plugins, but yeah, process beats any particular LLM.

that said, Claude Code is more mature and has a big ecosystem.

I don’t love what OpenAI is doing with an app store walled garden, although that has nothing to do with Codex’s efficiency or code quality.

DRAM and NAND memory prices have jumped more than 600% over the last one year for consumer applications, driven by surging demand from higher-margin AI server segments. Memory contribution to the router bill of materials has gone up to more than 20% share from 3% a year ago. - Counterpoint Research

Security experts identify indirect prompt injection as an architectural vulnerability where attackers hide commands in data sources, recommending sandboxing, strict access controls, and multi-model consensus to prevent unauthorized AI agent actions. - Reddit

LLM agent costs scale quadratically because every turn requires reading the entire conversation history, with cache reads accounting for up to 87% of total expenses and 50% of API costs at 27,500 tokens. - exe.dev

FT Alphaville's The Economics Show examines whether intense domestic competition in China, known as involution, will affect the robotics and artificial intelligence sectors as it has the electric vehicle market. - Financial Times

Qwen 3.5 claimed to match top proprietary models; DeepSeek update expected soon. - Qwen

Alibaba's Qwen AI model exhibits pro-China bias on human rights - China Media Project

Robotera's L7 humanoid robot performed a traditional Chinese sword dance using 55 degrees of freedom to demonstrate advanced motor control and real-time AI coordination for industrial and service applications. - Interesting Engineering

Bioengineer César de la Fuente uses the ApexOracle AI model to discover antimicrobial peptides from venom and extinct species, following successful animal trials against World Health Organization priority bacteria. - MIT Technology Review

San Francisco startup The Biological Computing Company raised $25 million to develop a platform integrating living neurons with AI models, claiming the technology processes data five times faster than silicon chips with lower power consumption. - Tom's Hardware

A literal second brain in a petri dish in a Matrix in a data center…

More than 700 US business schools have integrated AI into their curricula, though the absence of standardized benchmarks has prompted new initiatives to establish ethical and capability frameworks. - Financial Times

KPMG fined a partner for using artificial intelligence to complete an internal AI competency exam, marking one of several recent instances where staff used the technology to bypass internal testing requirements. - Financial Times

Maybe they were the only one that understood the assignment

Ars Technica retracted an article containing fabricated quotes generated by ChatGPT after a staff writer accidentally included the AI's paraphrased output as direct quotations. - 404 Media

Nancy Cartwright, the voice of Bart Simpson, opposed AI replacement in voice acting, citing a lack of human emotion, while Disney explores AI tools following a $1 billion licensing deal with OpenAI. - Deadline

don’t have a cow man

Professional wrestler El Grande Americano criticized his rival's use of AI-generated vignettes and voice synthesis in Lucha Libre AAA promotions, labeling the AI-filtered depiction of the Mexican flag a sign of disrespect to his culture. - Cageside Seats

Sentai, a UK-developed AI voice companion, provides seniors with medication reminders, appointment tracking, and camera-free fall detection for a £29.50 monthly subscription to support independent living. - TechRadar

Casio released Moflin, a $429 AI-powered social robot designed to simulate emotions and develop a unique personality for consumers in Japan and South Korea who cannot keep live animals. - The Verge

Follow the latest AI headlines via druce.ai on Bluesky