- 1Safety
AI-assisted op-eds used in anti‑Waymo campaign, often without disclosure
Opponents of Waymo have placed numerous opinion pieces in local outlets arguing against autonomous taxis, and many of those op-eds show high levels of AI-generated text according to the AI-detection service Pangram.
- 2Tools
How Stanford’s CS146S Teaches AI-Centric Software Engineering — Mihail Eric on Monaco’s Approach
Stanford’s CS146S, taught by Mihail Eric, reframes software-engineering assignments for an AI-powered era: students must learn an unfamiliar open‑source repository, propose a change, and defend the patch to a potential maintainer.
- 3Research
Anthropic tests 'dreaming' memory consolidation to improve AI agents' long-term performance
Anthropic studied how its Claude language models can use idle-time "dreaming" to review, correct and restructure their memory files, improving performance on multi-session tasks.
- 4Industry
Shadow AI Widens as Firms Lack Visibility and Governance
A PwC Hungary survey finds that a large share of workplace AI use falls outside corporate control: 67% of employees who use AI for work also rely on personal or free tools, and 42% say their organisation has no internal AI policy or they are unaware of one.
- 5Regulation
Clean-room coding, AI and copyright: renewed disputes over software authorship
The rise of generative AI has revived long-standing legal and technical debates about when new software crosses the line into infringement.
- 6Business
Anthropic IPO outlook clouds as valuation, revenue targets and investor confidence come under pressure
Anthropic’s planned IPO faces renewed scrutiny after reports show the company reached an $80 billion annualized run rate in October but still trails a $100 billion year‑end target.
- 7Model launches
Underdog debuts 31.7 MB Woof text‑to‑speech demo playable in the browser
Underdog released a browser game demonstrating Woof, a 31.7 MB text‑to‑speech model that runs locally without internet and contains about 31 million parameters.
- 8Safety
Nvidia board launches independent probe after reports linking CEO Jensen Huang to Chinese spy operation
Nvidia's board has opened an independent investigation after reports that CEO Jensen Huang was targeted by a Chinese spy operation and that he had an extramarital relationship with a former Megaspeed executive.
- 9Tools
Anthropic rolls out modification to the reasoning summarizer per account
Anthropic is rolling out the change to the reasoning summarizer per account in recent days: according to user reports the system’s performance has not deteriorated, and the development aims to…
- 10Research
How AI Is Changing Scientific Modeling
Jakob Macke, Daniel Gedon and Stefan Wahl recently demonstrate how artificial intelligence is beginning to automate and accelerate the writing, running, fitting and revision of scientific models;…
- 11Industry
Starbucks tests drive‑thru AI recognition to personalise menus — privacy concerns raised
Starbucks is exploring a drive‑thru system that would recognise consenting repeat customers on arrival and show personalised digital menus, CEO Brian Niccol said.
- 12Regulation
Parents sue Google alleging search AI contributed to 13-year-old's suicide attempt
The parents of a 13-year-old filed a lawsuit against Google on November 2 in a California court, alleging that the company's search engine AI contributed to their daughter's serious mental-health crisis.
- 13Business
Google takes over unfinished Wyoming data center project from Crusoe, pledges $17 billion investment
Crusoe sold an incomplete data center development in southeastern Wyoming to Google, which cited concerns over cost and schedule under Crusoe’s management.
- 14Model launches
Thinking Machines releases Inkling 2 Small: a 594B multimodal MoE with fast pretraining and planned open weights
Thinking Machines launched Inkling 2 Small, a multimodal mixture-of-experts model (594B total / 17B active) pretrained in under three days on native NVFP4 and offering large context support and native vision.
- 15Safety
OpenAI and the Financial Markets Forum launch a new framework to assess risks of AI exploitations
OpenAI collaborated with the Financial Markets Forum and its member organizations to develop a new framework that helps identify when the circumvention of an AI system’s defenses could lead to serious cyber harms, and to prioritize stronger protective measures.
