Australian universities set different rules for AI in marking
Newcastle, Melbourne and UNSW set different boundaries for AI-assisted marking. Their rules also differ on disclosure, opt-outs and academic responsibility.
Explore our reporting on how AI and robotics change work, business and the world.
Newcastle, Melbourne and UNSW set different boundaries for AI-assisted marking. Their rules also differ on disclosure, opt-outs and academic responsibility.
Atria Dawn's team studied 769 task records from its own model development. Agents often proposed methods and revised outputs, while participants reported making most final choices. The findings describe one project.
At Fellows Forum, Boris Power estimated that 80–90% of OpenAI research targets later GPT generations. The figure is his view, without a published allocation breakdown or GPT-7/8 roadmap.
NVIDIA's September 23 open-weight diarization model labels up to eight speakers in recorded or streaming audio. Its shortest recommended 0.32-second setting measures buffered input, while accuracy figures come from NVIDIA's own tests.
Caltech and Stanford researchers show a Unitree G1 using a remembered kitchen map and reusable skills. The release has selected demonstrations but no controlled results, full paper or robot code.
Newly public book-copyright filings dispute whether acquiring books and training AI require separate legal analysis. OpenAI and Microsoft contest the authors’ case.
Torridge says Xlinks has not submitted a formal application for its proposed Devon AI campus. The council distinguishes early planning advice from the public statutory process.
A September research report describes DSec, DeepSeek's sandbox platform for agent training, including multiple backends, composable environments and author-reported deployment scale.
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 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.
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.
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 Democratic officials have taken different routes on AI oversight, data-center construction and worker support. State orders are in effect; federal measures remain proposals.
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.
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 its Bodleian digitisation pilot contributes training data to OpenAI. Public project records detail scanning and transcription, but not the material or models involved.
A practical guide to using Rive CLI and RML with a coding agent to build, inspect and test interactive graphics, including account, signing and export limits.
A public Claude Code skill set combines Stockfish evaluations, player notes and model-written explanations in a reviewable chess post-mortem. We trace its documented workflow and limits.