Bittensor subnet news

KubeTEE AI@KubeTEEAI

Running workloads in TEE Confidential Computing is part of the security; continuous attestation with a measurement breaker policy protects against code injection. I guess they miss the important part.

Quoting @near_ai

Update: On November 8, an attacker used a vulnerability in SGLang, the inference engine we run, to access part of our infrastructure. We’ve closed the access and fixed it. Affected users have been contacted. If you haven’t heard from us, no action is needed. near.ai/blog/security-incident-report-sglang-vu…

Full post on TaosisOpen on X ↗

KubeTEE AI@KubeTEEAI

Every host is always updated, monitored, and audited in KubeTEE with the best practices for hardening the OS to keep all vectors of attack SAFU

Quoting @lnxsec

Detection often begins with something breaking downstream, not with the exploit itself. Phala says an automated certificate/CAA reconciliation failure detected the incident roughly 17.5 hours after the first malicious action on the affected host. The signal came from a control-plane inconsistency after gateway settings had been changed. That is a useful security-operations pattern. Reconciliation, drift detection, and expected-state checks can reveal compromise even when the original exploit does not produce an obvious alert. Linux and cloud teams sometimes treat these checks as reliability tooling. They can also be integrity sensors. **In practical terms, it is a good time to:** - alert on unexpected changes to firewall, DNS, ACME, gateway, and certificate-management configuration - compare desired state in version control with active host and service configuration - retain the actor, source, timestamp, and before/after values for administrative changes - test whether reconciliat

Full post on TaosisOpen on X ↗

KubeTEE AI@KubeTEEAI

The main reason is privacy; fixing your security flaws needs to be kept private, not leaked anywhere. KubeTEE provides a hardware-encrypted, attested Trusted Execution Environment with Confidential Computing technology.

Quoting @guilleflorvs

BREAKING: Revolut moved ~2,000 engineers onto self-hosted GLM for coding and pushed Claude Code and Codex to the exception list. Per David Abitbol, who leads AI ops there: 56 Nvidia B200s, a blended 30 cents per million tokens, roughly a tenth of what they paid Anthropic and OpenAI. Engineering lead time improved 70%. Here's the part that matters: the gains showed up when tokens stopped being rationed. Engineers quit saving AI for the hard tickets and pointed it at everything. Frontier models become the exception, open weights become the default. That's a pricing problem for every lab heading to IPO. h/t @OliverMolander Would you move your team off frontier APIs if the bill dropped 90%?

Full post on TaosisOpen on X ↗

KubeTEE AI@KubeTEEAI

This is exactly what KubeTEE SLO (Service Level Objectives) aims for, with TEE Confidential Computing included.

Quoting @OpenRouter

We just shipped our biggest improvement to our provider router since the founding of OpenRouter. Before, we prioritized mostly based on price. The new router now rewards the best quality inference for the actual cost of a request, so requests stay cheap but are also fast and reliable. Providers that deliver more value automatically win more traffic, so every provider has a reason to keep improving. And every time a provider gets better, so does your inference on OpenRouter.

Full post on TaosisOpen on X ↗

KubeTEE AI@KubeTEEAI

In KubeTEE, when a pod starts, it does an attestation that Confidential Computing is active and trusted by Intel Trust Authority. Only then does the pod receive the secrets from the KBS that gates it.

Quoting @IntCyberDigest

‼️ UPDATE: Hacker "k0t" breached 924 more Fortinet firewalls than first known, Hudson Rock says. They hold 40,573 cleartext secrets, including 1,576 Active Directory bind passwords.

Full post on TaosisOpen on X ↗

KubeTEE AI@KubeTEEAI

Confidential Computing TEE attestation would kill any code changes. KubeTEE policies protect against this.

Quoting @TheHackersNews

🚨 PHP2Shell exploited in real-world attacks. An attacker chained PHP2Shell with a Linux kernel flaw to gain root access to 25 servers across four countries, including three in a U.S. government-affiliated environment. The intrusion extended into the internal network, where the attacker targeted Active Directory. Suspected China nexus. Signs of AI-assisted tooling. Read the full report: thehackernews.com/2026/10/php2shell-attack-root…

Full post on TaosisOpen on X ↗

KubeTEE AI@KubeTEEAI

BLACK OCTOBER That is the beauty of TEE attestation policies (If measure change = shutdown) Guarantee that the code that should be running is not changed, or else TRIGGER THE BRAKER OFF.

Quoting @cgtwts

the cybersecurity industry has had a generational disaster of a month. they’re calling it black october for a reason. > Oct 5: SubQuery npm package backdoored > Oct 7: IDCF Cloud hit by ransomware, affecting 495 organizations > Oct 8: TensorLake publishes malicious npm package > Oct 14–15: Zeabur infrastructure breached > Oct 15: GhostAction targets GitHub repos with malicious workflows > Oct 15–16: Rank Math PRO update delivers malware > Oct 16: ConnectWise cloud backup identities compromised > Oct 17–18: TeamUpdraft customer data compromised > Oct 18–19: OpenCode exposes AI API keys > Oct 19: Nikkei reports 1.83M records exposed > October 19: OVHcloud CLI binaries reportedly backdoored > Oct 20–21: PHP critical RCE vulnerabilities disclosed > Oct 21: LiteLLM ships malicious PyPI + Docker releases > Oct 21: Railway reports potential exposure of customer secrets > Oct 21: eMudhra certificate authority keys compromised > Oct 21: BIND 9 security vulnerabilities disclosed > October 21:

Full post on TaosisOpen on X ↗

KubeTEE AI@KubeTEEAI

Provided by KubeTEE with Minimax-H3 in TEE

Quoting @say_gm_

Video is live on SayGM 🧵 A child's crayon dragon takes flight. One sentence becomes a stop-motion sunrise. Two still images become a single clip. Here's how each one was made, so you can make your own.

Full post on TaosisOpen on X ↗

KubeTEE AI@KubeTEEAI

Our Confidential Computing tech stacks are NOT vulnerable to the Vghost attack surface This is the advantage of TEE

Quoting @TheHackersNews

🛑 A Linux kernel flaw can let an untrusted VM hijack its host's QEMU process in specific setups. Dubbed Vghost (CVE-2026-74580), the AI-assisted discovery also enables local root escalation for users with vhost device access. Read: thehackernews.com/2026/10/ai-found-linux-kernel…

Full post on TaosisOpen on X ↗