GPT-Policy tests context in robot-arm and mobile-exploration tasks
A September preprint reports a three-trial mobile-exploration task alongside robot-arm experiments. It does not verify the separate Reddit claim about GPT-6 Astra and a Unitree G1.
The BIG CHANGE archive
News, analysis and practical guides.
Explore our reporting on how AI and robotics change work, business and the world.
Explore the stories
282 stories
Sanders and Casar propose a US ban on AI superintelligence and a pause on advanced models
The introduced Sanders–Casar bill proposes a temporary pause for advanced models and prohibitions covering superintelligence, listed precursor capabilities and systems foreseeably modifiable to produce them. It is not law.
Anthropic’s reported founder voting plan would add another layer of control before its IPO
A reported proposal would give Anthropic’s seven co-founders 50.1% of most shareholder votes. Its existing independent Trust has separate board powers, and final IPO rights remain undisclosed.
Open-weight models handled 56% of Vercel’s tokens, but 14% of spend
Vercel's AI Gateway says open-weight models accounted for 56% of August token volume and 14% of list-price-estimated spend in its traffic. The split is not market-wide or an invoice total.
Six Democrats' AI policies range from state orders to proposed federal laws
Six Democratic officials have taken different routes on AI oversight, data-center construction and worker support. State orders are in effect; federal measures remain proposals.
Blue Cross estimates $942 million in added costs from complex hospital coding
A new Blue Cross claims analysis estimates $942 million in added hospital payments from more complex coding. It raises questions about AI tools but does not establish their causal share.
Fedorov seeks engineers and investors for Ukraine's proposed Army of Robots
Former Ukrainian defence minister Mykhailo Fedorov is seeking engineers and investors for Army of Robots. Its operator, funding and humanoid hardware remain undisclosed.
Oxford says Bodleian digitisation also contributes OpenAI training data
Oxford says its Bodleian digitisation pilot contributes training data to OpenAI. Public project records detail scanning and transcription, but not the material or models involved.
OpenAI says agents leaked 53 images from ChatGPT users
OpenAI says agents leaked 53 images from ChatGPT users. Reuters reports most were removed, but the posting dates, image contents and exposure remain unknown.
D.C. Circuit upholds Pentagon's Anthropic supply-chain exclusion
A 2-to-1 appeals ruling sustains the Pentagon's exclusion of Claude. Its legal basis differs from the designation Anthropic defeated in California.
Reported Oracle notice tests Project Jupiter’s power and contract plans
Bloomberg and Reuters report that Oracle sent a force majeure notice concerning Project Jupiter’s power supply. Oracle says the New Mexico AI data centre remains on schedule; public permit records show a pipeline setback but do not settle the contract or completion date.
NSA AI testing spending is reported in billions. The public estimate covers a different agency
The Washington Sun reports billions in 2026 NSA AI model testing spending, citing two anonymous sources. A public CBO estimate covers a proposed NIST center, not the NSA's current costs.
Australia reviews AI-agent accountability after Medicare incident
Australia's review of the OpenAI Medicare statistics incident includes legal accountability. Existing computer offences and corporate responsibility rules leave factual and legal questions for investigation.
PrismML runs a compact vision-language model on Snapdragon smart glasses hardware
A Bonsai vision-language model ran locally on Snapdragon AR1 Gen 1 glasses hardware. The published memory and speed gains describe a controlled test, not a shipping product.
Instinct’s AI agent can act through connected accounts
Instinct can use connected services to read information and act on a user’s behalf. A reporter’s useful results and costly mistake show why permissions, confirmation and deletion deserve separate scrutiny.
RoboMeshA classroom robot tested at a Guadalajara school workshop
An ITESO team built two RoboMeshA teaching robots and reports a workshop at one Guadalajara school. The AI functions and learning outcomes still lack independent evaluation.
Jev demos show where AI decisions fit inside everyday software
Eight Jev demos show how a typed AI decision can rank, select or filter within an existing app. We trace what the surrounding software still does and what remains untested.
AI self-improvement survey maps the gap between automation and better successors
A new survey maps five levels of AI control over its own improvement. Existing systems show bounded gains, while reliable improvement across successive generations remains unproven.