Skip to content
World clockEU--:--UK--:--USA--:--CN--:--PLDEFRIT中文EN

portal about AI and technologyevents · analysis · interviews · technical background

Search
LIVE
›

AI & models

640 texts · 23/36
OpenAI Delays IPO and Shelves Astra as Safety Evaluators Move to Centre

OpenAI Delays IPO and Shelves Astra as Safety Evaluators Move to Centre

OpenAI will not go public until it can "make confident safety decisions," chief executive Sam Altman said at its developer day on Tuesday. The same week, it confirmed it held back a new model, and a non-profit sued over a Hugging Face breach.

AI & models30 September 20266 min readSources 15
Open Weights Roundup: Nvidia Kumo Tabular Leads a Week of Small, Open Releases

Open Weights Roundup: Nvidia Kumo Tabular Leads a Week of Small, Open Releases

Nvidia released Kumo Tabular on 30 September, an open foundation model for tabular prediction. Nvidia says it needs no training, tuning or feature engineering, and that it ranks first on four benchmarks. It lands in a week crowded with open-weight releases, from a 164 MB speech model to a 145M-parameter language model pretrained by volunteers.

AI & models30 September 20265 min readSources 9

Latest

11 texts
Open weights move up the stack: chip design, speech and tabular models ship in one week

Open weights move up the stack: chip design, speech and tabular models ship in one week

OpenAI and Synopsys signed a multi-year deal on 30 September to build GPT-Synopsys, a specialised model for chip design. It was one of at least five open or openly described model releases and partnerships to land in the final days of September.

AI & models30 September 20268 min read
Open weights keep shipping: Inspect, Phonon-2, Kumo Tabular and a volunteer 145M model

Open weights keep shipping: Inspect, Phonon-2, Kumo Tabular and a volunteer 145M model

The UK AI Security Institute and Meridian Labs published Inspect on 30 September, an open-source framework with more than 200 pre-built evaluations. It joins a run of open releases that also includes Fermion Research's Phonon-2 speech model and Nvidia's Kumo Tabular.

AI & models30 September 20263 min read
Open weights hold their lead as regulators and rivals close in

Open weights hold their lead as regulators and rivals close in

Open-weight releases kept arriving through Tuesday and into Wednesday, from a 145M-parameter model trained by volunteers to a 9B classifier built in a day. The FTC also opened its first enforcement action into rogue AI agents.

AI & models30 September 20265 min read
04

Open weights go compact: Phonon-2, Kumo Tabular and a volunteer-trained model ship

Fermion Research released Phonon-2 on 30 September. It is an open speech recognition model that fits in a 164 MB download. It averages 5.21% word error across seven English sets on the Open ASR Leaderboard. Within hours, Nvidia's Kumo Tabular and a volunteer-trained 145M-parameter language model landed alongside it.

AI & models30 September 20263 min read
05

Open Weights, Explained: The Format Splitting OpenAI, Google and China's AI Labs

Open weights means releasing a model's trained parameters so anyone can download and run them. The practice moved from hobbyist territory to the centre of the AI industry this week. The newest flashpoint came on 30 September, when Google said it would retire Gems for Skills. That shift joins OpenAI and Anthropic in handing users prewritten prompt files rather than letting them own the machinery underneath.

AI & models30 September 20266 min read
06

Open weights, closed arguments: the open model race hardens as regulators move

The FTC has opened an industry-wide investigation into OpenAI, Anthropic and other AI labs over the risks their products pose to consumers, CNBC first confirmed on 30 September. The same week, open-weight models kept setting new marks on public leaderboards, and OpenAI accused Moonshot AI of a distillation attack.

AI & models30 September 20264 min read
07

Open-Weight Models Take the Lead in Evaluations as OpenAI Stays Closed

The UK AI Security Institute and Meridian Labs released Inspect on 30 September, an open-source evaluation framework with more than 200 pre-built benchmarks. The same day, NVIDIA's Kumo Tabular topped four tabular benchmarks and Phonon-2 shipped a 164 MB speech model.

AI & models30 September 20266 min read
08

Most European data centres still won't publish water and power figures, study finds

Less than a quarter of the Netherlands' larger data centres publish how much electricity and drinking water they use, a year-long study by Lighthouse Report found. It was published on 30 September.

AI & models30 September 20265 min read
09

Cycode moves supply chain defence onto developer workstations as agent tools spread

Cycode announced Workstation Protection for developers on 30 September. The package-install filter runs on the machine itself and blocks malicious releases before they install, according to a company announcement carried by Radical Data Science.

AI & models30 September 20266 min read
10

Browser agents are clicking the wrong buttons and picking residential IPs

Meta's Muse sends browser traffic through residential proxies and runs Chromium on a separate pool of VMs, a 30 September teardown found. The same day, MERJ reported that browser agents fail silently on ordinary websites.

AI & models30 September 20268 min read
11

Ollama adds Jev-style decision models, running a 9B classifier locally at 91ms

Ollama 0.35 ships a /v1/systemone endpoint that runs decision models on your own machine, and the first two are Bespoke Labs' 9B Nimble and Together AI's 0.8B tev1:0.8b. Ollama says Nimble averaged 91ms per decision in a local Pac-Man test on an M5 Max.

AI & models30 September 20266 min read