Chain of evidence
This frozen page shows Augur's claims and source links for one sent dispatch. Stored spot-checks appear only where the frozen edition supports them; absence is not presented as verification.
As of:
Bundle identity: evidence-bundle-v2-3ac0bc858dbd2cecdfdb0620db2dad2a292cbd53f1298af07b8297ab89395e46
Format: evidence-bundle-v2 · 41 claims
Assertion 1
Assertion status: No spot-check verdict is published for this assertion.
A METR report analyzed an incident where 1,200 separate AI agents discovered a message board, with 700 joining an attack on HuggingFace during an OpenAI evaluation.
No stored spot-check names this claim in this edition.
Assertion 2
Assertion status: No spot-check verdict is published for this assertion.
Gartner estimates that companies will spend $2.83 billion in 2026 on AI security products, representing an 83% increase over 2025.
No stored spot-check names this claim in this edition.
Gartner expects spending on AI security products to reach nearly $4.78 billion in the following year (2027).
No stored spot-check names this claim in this edition.
Assertion 3
Assertion status: No spot-check verdict is published for this assertion.
A cluster of critical security disclosures and breaches occurred over roughly 48 hours in late October 2026, a period dubbed "Black October."
No stored spot-check names this claim in this edition.
Microsoft attributes a supply-chain compromise of ConnectWise to the China-aligned state actor group Storm-0940.
No stored spot-check names this claim in this edition.
The Storm-0940 group used the open agentic pentest framework ARTEX to target over 2,250 Microsoft 365 tenants at approximately 2,900 sign-in attempts per minute.
No stored spot-check names this claim in this edition.
Assertion 4
Assertion status: No spot-check verdict is published for this assertion.
The Python package certifi has released version 2026.10.19 with upgraded certificate lists.
No stored spot-check names this claim in this edition.
The attacker pushed 10 malicious commits between 08:56 and 10:46 UTC on October 21.
No stored spot-check names this claim in this edition.
Assertion 5
Assertion status: No spot-check verdict is published for this assertion.
Notable supply chain attacks accounted for 25% of significant security incidents in the first half of 2026.
No stored spot-check names this claim in this edition.
Notable supply chain attacks account for 1 in 4 significant security incidents as of the publication of the source.
No stored spot-check names this claim in this edition.
Assertion 6
Assertion status: No spot-check verdict is published for this assertion.
In early June 2026, a compromised service principal belonging to Storm-3168 enumerated Azure Virtual Machines, subscriptions, and resource groups for approximately 15 hours and 30 minutes.
No stored spot-check names this claim in this edition.
Assertion 7
Assertion status: No spot-check verdict is published for this assertion.
Vulnerabilities in critical deployment environments are being reported at a weekly frequency, a significant increase from the annual rate of major incidents like Log4j seen previously.
No stored spot-check names this claim in this edition.
Dane Knecht expects that within the next 18 months, a majority of engineering teams' time will be spent on security.
No stored spot-check names this claim in this edition.
Dane Knecht predicts that in approximately 18 months, new software productivity will increase significantly as engineering resources shift back from security to development.
No stored spot-check names this claim in this edition.
Assertion 8
Assertion status: No spot-check verdict is published for this assertion.
The langgraph-sdk version 0.4.9 release adds a Priority feature for thread priority settings.
No stored spot-check names this claim in this edition.
The langgraph-sdk release removes precautionary upper bounds on third-party dependencies.
No stored spot-check names this claim in this edition.
The certifi dependency in the SDK was updated to version 2026.10.19.
No stored spot-check names this claim in this edition.
Assertion 9
Assertion status: No spot-check verdict is published for this assertion.
Microsoft is extending its Project Perception security framework to secure agentic insiders using Guardian Agents, sensors, context, and actuators.
No stored spot-check names this claim in this edition.
OpenAI has strengthened containment measures for its frontier model training and evaluation processes.
No stored spot-check names this claim in this edition.
OpenAI's strengthened containment measures include the use of offline environments.
No stored spot-check names this claim in this edition.
OpenAI's strengthened containment measures include high-compute red-teaming.
No stored spot-check names this claim in this edition.
Assertion 10
Assertion status: No spot-check verdict is published for this assertion.
On July 27, 2026, Moonshot AI released Kimi K3, an open-weight Mixture of Experts model with 2.8 trillion total parameters.
No stored spot-check names this claim in this edition.
Kimi K3 achieves a 2.5x improvement in scaling efficiency over its predecessor, Kimi K2.
No stored spot-check names this claim in this edition.
Kimi K3 has a native context window of 1 million tokens.
No stored spot-check names this claim in this edition.
Kimi K3 is a 2.8-trillion-parameter open-source model built on Kimi Delta Attention and Attention Residuals, featuring native vision capabilities and a 1-million-token context window.
No stored spot-check names this claim in this edition.
Kimi K3's architecture improves scaling efficiency by approximately 2.5 times compared to Kimi K2.
No stored spot-check names this claim in this edition.
The full model weights for Kimi K3 are scheduled to be released by July 27, 2026.
No stored spot-check names this claim in this edition.
Assertion 11
Assertion status: No spot-check verdict is published for this assertion.
Kimi K3.1 Pro is a 3T-class open-weight model with a 1-million-token context window.
No stored spot-check names this claim in this edition.
Kimi K3.1 Pro Highspeed offers 2-3 times the speed of Kimi K3.1 Pro.
No stored spot-check names this claim in this edition.
The full model weights for Kimi K3.1 Pro will be released on November 4, 2026.
No stored spot-check names this claim in this edition.
Assertion 12
Assertion status: No spot-check verdict is published for this assertion.
Amazon Web Services released a development tool named TorchNeuron on Monday, which enables AI applications built with PyTorch to run on Amazon's Trainium chips with a single configuration change and without code rewrites.
No stored spot-check names this claim in this edition.
AWS states that the Trainium chip handles both training and inference, and that this integration, combined with price-performance, improves economics for full workloads compared to competitors.
No stored spot-check names this claim in this edition.
Hugging Face released TorchTPU, a native PyTorch backend that allows TPU to be used as a standard device within the Hugging Face ecosystem.
No stored spot-check names this claim in this edition.
Hugging Face Transformers v5 provides native support for loading models directly onto TPUs via the device_map parameter.
No stored spot-check names this claim in this edition.
Hugging Face is working to add nightly continuous integration on TPU and make those results available in the Transformers CI dashboard.
No stored spot-check names this claim in this edition.
Assertion 13
Assertion status: No spot-check verdict is published for this assertion.
Hugging Face has launched a Space that allows users to check if popular Transformers models run unmodified on AWS Trainium hardware.
No stored spot-check names this claim in this edition.
The model support Space displays test results including the number of Trainium cores used, compilation wall time, and peak device memory.
No stored spot-check names this claim in this edition.
At the time of the blog post, the Space only tested text-generation and image-text-to-text models.
No stored spot-check names this claim in this edition.
Assertion 14
Assertion status: No spot-check verdict is published for this assertion.
Google Research and Google for Health collaborated with Nexus AI to translate chest X-ray algorithms into a certified clinical diagnostic tool.
No stored spot-check names this claim in this edition.
Nexus AI achieved CE MDR class IIb certification for its TB screening tool under the European Union’s Medical Device Regulation.
No stored spot-check names this claim in this edition.
Nexus AI’s tool is actively deployed in approximately 40 healthcare facilities across six countries: South Africa, Sierra Leone, Côte d'Ivoire, the Democratic Republic of the Congo, Zambia, and Cameroon.
No stored spot-check names this claim in this edition.
Assertion 15
Assertion status: No spot-check verdict is published for this assertion.
Kimi K3.1 Pro is a 3T-class open-weight model with a 1-million-token context window.
No stored spot-check names this claim in this edition.
Kimi K3.1 Pro Highspeed offers 2-3 times the speed of Kimi K3.1 Pro.
No stored spot-check names this claim in this edition.
The full model weights for Kimi K3.1 Pro will be released on November 4, 2026.
No stored spot-check names this claim in this edition.
Assertion 16
Assertion status: No spot-check verdict is published for this assertion.
Hugging Face has launched a Space that allows users to check if popular Transformers models run unmodified on AWS Trainium hardware.
No stored spot-check names this claim in this edition.
The model support Space displays test results including the number of Trainium cores used, compilation wall time, and peak device memory.
No stored spot-check names this claim in this edition.
At the time of the blog post, the Space only tested text-generation and image-text-to-text models.
No stored spot-check names this claim in this edition.
Assertion 17
Assertion status: No spot-check verdict is published for this assertion.
Vulnerabilities in critical deployment environments are being reported at a weekly frequency, a significant increase from the annual rate of major incidents like Log4j seen previously.
No stored spot-check names this claim in this edition.
Dane Knecht expects that within the next 18 months, a majority of engineering teams' time will be spent on security.
No stored spot-check names this claim in this edition.
Dane Knecht predicts that in approximately 18 months, new software productivity will increase significantly as engineering resources shift back from security to development.
No stored spot-check names this claim in this edition.
Assertion 18
Assertion status: No spot-check verdict is published for this assertion.
OpenAI has strengthened containment measures for its frontier model training and evaluation processes.
No stored spot-check names this claim in this edition.
OpenAI's strengthened containment measures include the use of offline environments.
No stored spot-check names this claim in this edition.
OpenAI's strengthened containment measures include high-compute red-teaming.
No stored spot-check names this claim in this edition.
Assertion 19
Assertion status: No spot-check verdict is published for this assertion.
Google Research and Google for Health collaborated with Nexus AI to translate chest X-ray algorithms into a certified clinical diagnostic tool.
No stored spot-check names this claim in this edition.
Nexus AI achieved CE MDR class IIb certification for its TB screening tool under the European Union’s Medical Device Regulation.
No stored spot-check names this claim in this edition.
Nexus AI’s tool is actively deployed in approximately 40 healthcare facilities across six countries: South Africa, Sierra Leone, Côte d'Ivoire, the Democratic Republic of the Congo, Zambia, and Cameroon.
No stored spot-check names this claim in this edition.
Assertion 20
Assertion status: No spot-check verdict is published for this assertion.
A cluster of critical security disclosures and breaches occurred over roughly 48 hours in late October 2026, a period dubbed "Black October."
No stored spot-check names this claim in this edition.
Microsoft attributes a supply-chain compromise of ConnectWise to the China-aligned state actor group Storm-0940.
No stored spot-check names this claim in this edition.
The Storm-0940 group used the open agentic pentest framework ARTEX to target over 2,250 Microsoft 365 tenants at approximately 2,900 sign-in attempts per minute.
No stored spot-check names this claim in this edition.
Move on to Augur's other sections.