Loading story…
GitLab has expanded GitLab Duo Self-Hosted to support AI models deployed through Microsoft Foundry in an organization’s selected Azure environment. The integration lets enterprises route Duo requests through their own AI Gateway and model endpoints instead of GitLab-managed model infrastructure, providing greater control over data flows, model deployment locations, and credentials. Supported model families include OpenAI GPT, Anthropic Claude, Meta Llama, and Mistral, with feature-level model selection available for individual Duo capabilities.
The self-hosted option shifts responsibility for operating the AI environment to customers. Organizations must provision sufficient model capacity, configure network connectivity and credentials, maintain availability and lifecycle management, and validate that their selected Microsoft Foundry models are compatible with GitLab Duo. The capability is aimed at enterprises with data-residency, governance, or infrastructure-control requirements for AI-assisted software development.
Track how attackers are adapting to this technology.
1 event from the most recent confirmed update back to the earliest known activity.
GitLab expanded GitLab Duo Self-Hosted to support AI models deployed through Microsoft Foundry in an organization's chosen Azure environment. The integration lets customers use self-hosted AI Gateway and model endpoints rather than GitLab-managed model infrastructure.
Vulnerabilities, threat actors, malware, products, organizations, and breaches Mallory has linked to this story.
Follow how adversaries are adapting to this technology, and where it touches your stack today.
2 references tracked. Mallory keeps watching after this page renders.
Map indicators from this story to your assets and identify affected systems in minutes.
Every observed campaign, victim, and pivot linked to actors named in this story.
Malware, exploits, and IOCs connected to the activity described here.
YARA, Sigma, and Snort rules deployed to your SIEM as soon as they’re published.
Get matching new stories delivered to your team as they break — not the next morning.
Ask questions about this story and take action on the answers.
Other stories Mallory is tracking that overlap on entities, sources, or topic labels.
OpenAI released **GPT-5.4**, positioning it as a capability and efficiency upgrade aimed at knowledge-work use cases, including improved visual understanding (supporting analysis of very large images) and a claimed reduction in factual errors versus prior versions. The model is being rolled out across *ChatGPT* (web and apps), *Codex*, and the API, with higher-tier variants (e.g., “Thinking” and “Pro”) available to paid and enterprise/education offerings; the coverage also notes competitive pressure from **Anthropic** and heightened scrutiny around U.S. government/military AI use. Separately, OpenAI is reportedly developing an internal (and potentially customer-facing) alternative to **GitHub** after repeated GitHub outages disrupted engineering workflows, with the reliability issues tied to GitHub’s ongoing migration from a legacy data center to **Microsoft Azure** and associated configuration/underlying cloud incidents. In parallel, U.S. policymakers are drafting stricter, worldwide export controls for advanced AI accelerators from U.S. vendors (including **Nvidia** and **AMD**), proposing a tiered licensing regime with increasing compliance obligations (e.g., transparency requirements and potential inspections) as deployment scale grows—measures that could materially affect large AI infrastructure buildouts by major operators.
Jun 29, 2026
OpenAI announced **GPT-5.3-Codex**, positioning it as more than a code-generation model and expanding its availability across a new macOS desktop app plus CLI, IDE extension, and web interface, with API access described as forthcoming. OpenAI and third-party coverage said the model is intended to support broader software-lifecycle work (e.g., debugging, deployments, monitoring, tests, and documentation) and highlighted benchmark gains over prior Codex versions; reporting also pushed back on claims that “Codex built itself,” characterizing OpenAI’s statement as the model being *instrumental* in its own development rather than fully autonomous. In parallel, OpenAI unveiled **Frontier**, an enterprise framework aimed at building, deploying, and managing AI agents, including promised agent security features and an approach modeled on Palantir-style **forward-deployed engineers** working alongside customer teams. Separately, *MintMCP* launched an enterprise governance platform focused on deploying, monitoring, and securing AI agents and MCP servers, emphasizing audit trails, policy enforcement, observability/guardrails, and centralized access controls to reduce risks from privileged agents (e.g., credential exposure and data exfiltration). Other items in the set were not tied to these launches, including a general CIO column on SMB IT “quick fixes” creating long-term risk and a corporate internal AI learning program announcement from Hancom.
Jun 29, 2026
AWS expanded **Security Hub** beyond its own cloud to add native visibility into **Microsoft Azure**, marking the service’s first direct monitoring of infrastructure outside AWS. The update lets Security Hub discover Azure virtual machines, container images, Function Apps, and identities, then evaluate them for misconfigurations, internet exposure, and vulnerable software. AWS said the Azure assessments align with the **CIS Microsoft Azure Foundations Benchmark**, giving organizations a standardized baseline for multicloud posture management. AWS also introduced new AI-focused security capabilities, including **GuardDuty AI Protection** for Amazon Bedrock and SageMaker and AI-powered investigations in Security Hub. The AI protections are designed to detect threats such as anomalous model use, prompt injection attempts, and cost-harvesting abuse tied to stolen credentials, while the investigations feature uses up to 90 days of related activity to generate confidence-scored findings, **MITRE ATT&CK** mappings, and remediation guidance. AWS said the expansion is aimed at unifying security operations across multicloud environments while improving visibility into both cloud infrastructure and AI assets.
Jul 14, 2026