Company
Scale AI is a data-labeling and AI infrastructure company that provides training data, evaluation, and benchmarking services for machine learning and AI models.
NVIDIA's open Nemotron 3.5 Super VL is a 120B mixture-of-experts model that activates 12B parameters per token for text, image, document and video work. NVIDIA's post still places it in a multi-model system, so consolidation covers one tier.
Reality
Refugees in Kenya's Kakuma camp, barred from legal work, are doing tech tasks for discretionary rewards of up to $500, the Guardian reported. Companies that buy crowd-sourced data and research work now have a labour question to put to their vendors.
Reality
Scale AI and Elorian's Humanity's Sixth Sense benchmark scores the best multimodal model at 53.6% of tasks, against 93.1% for human participants. Teams building on multimodal models should test spatial and social inference on their own footage before relying on it.
Reality
Three Axiom engineers had OpenAI's GPT-6 Astra steer a 2024 Toyota Corolla to an In-N-Out take-out window with no prior coaching. The rig was a chat interface wired through a server to windscreen cameras and the power steering, so the text model's output went to the car's steering.
Reality
Nvidia on Monday launched an AI agent safety platform whose Sentry watchdog runs on its BlueField-4 data processors. The design assumes agents will work around their limits, so its strongest enforcement runs on hardware only Nvidia makes.
Perspective Coverage
9 publishersReality
Halluminate raised a $30 million Series A led by Oak HC/FT to train AI agents on deal diligence, where the best of seven frontier models averaged 51%. Its customers are AI labs, so the round is a bet that model makers keep paying a nine-person firm for finance-specific training.
Perspective Coverage
4 publishersReality
David AI's 21-language benchmark found speech systems' error rates spread an average of 25.4 points in five Indic languages, against 3.7 in the tightest five. In those languages, choosing a system is a 25-point decision that an English score can make look like a close call.
Reality
NVIDIA's Open Agent Safety Platform, launched September 28, makes its OpenShell runtime available now while its Sentry hardware watchdog has no release date. Sentry's promised millisecond stop has no independent test yet, so the hardware half rests on NVIDIA's word.
Perspective Coverage
4 publishersReality
Forbes puts the 18-month-old training-data startup at $3.2bn against the $300m it carried five months ago. Most of that move is multiple rather than growth, and the round has not yet closed or named a lead.
Reality
Alexandr Wang says the new model costs developers no more than 1.2 and finishes the same work on about a quarter fewer tokens. That is a real saving on high-volume code generation and a rounding error most other places.
Perspective Coverage
3 publishersReality
Meta's fourth Muse Spark in five months gains four points on Artificial Analysis' Intelligence Index, almost entirely in agentic tests, while two scores fall and the tier Meta benchmarked against rivals is still a limited preview.
Perspective Coverage
6 publishersReality
Researchers counted at least 40,000 people in Nairobi writing other people's coursework early this decade. The trade is close to gone, and the work that still pays is rewriting model output by hand until the software cannot tell.
Perspective Coverage
3 publishersReality
Insight Partners and S32 led $350 million into a business that ships graded tasks, rubrics and reinforcement-learning environments to frontier labs. The price works out near ten times a company-reported run-rate.
Reality
Alexandr Wang, Meta's chief AI officer, has published more than 300 posts on X since Muse launched on September 8, announcing partner integrations and amplifying user savings claims, many of them from Meta's own employees.
Reality
Bain Capital Ventures led a $25 million seed into an AI-agent benefits brokerage selling to employers with one to 500 workers. Its claim of 20% savings gets a first real test inside one enrolment season.
Perspective Coverage
3 publishersReality
The company says a 20-person account earns a brokerage a fraction of an enterprise account for much the same work, and it is putting AI agents on that paperwork while keeping a licensed human advisor on every client.
Reality
Twenty-two provincial innovation centres opened within 14 months, and Wuhan alone logs 24,000 training entries a day. But Beijing's Shijingshan centre dropped its robot partner over weak data sales, so subsidy is still paying the bill.
Reality
Artificial Analysis now reports how often a provider or model refuses coding-benchmark work on safety grounds. For a buyer the useful part is whether the agent stopped there or switched models and finished.
Reality
The Justice Department told a New York federal court that licensing would make a robust AI industry harder to build. Vivek Shah, whose company is a plaintiff, answers with OpenAI's own compute forecast and the music industry's royalty bill.
Reality
Mecka AI pays people to wear sensors while they make coffee or fix cars, and two people told TechCrunch that Sequoia is leading a round at about $500 million, three months after a $60 million round. The terms are not final.
Reality
Build · September 2, 2026 · Confirmed2 publishers