News for today:
- [HackerOne] OpenAI models available in the H1 Platform. HackerOne’s H1 Platform products will feature OpenAI’s frontier GPT cyber models to assist with finding and fixing vulnerabilities faster; usage is optional and customer-controlled. The H1 Platform multi-model harness can use OpenAI’s frontier cyber models through the Daybreak Defense Network to broaden coverage, validate outputs, and trace findings to root cause. Security teams receive each finding with a developer-ready fix plan designed to cut overall time to remediate. HackerOne
- [Cloudflare] Ditto. Cloudflare’s doing something similar with its new Vulnerability Discovery and Remediation service; frontier models from OpenAI to enable deep code investigation and automated patch generation. By natively integrating AI code analysis directly across its unified platform—spanning Web Assets, WAF, and Workers Observability—Cloudflare enables organizations to move from isolated vulnerability findings toward context-aware prioritization and remediation. Available in early release. Cloudflare
- [SentinelOne] Ditto. SentinelOne is also doing something similar: offering a service that combines OpenAI’s frontier models with its own offensive and defensive security experts. In addition to the new models, the expanded Wayfinder Frontier AI Services will deliver two new capability areas, both reinforced by shared malware analysis, detection, and validation: (1) AI-powered code risk analysis, and (2) AI-enabled compromise assessment. SentinelOne
- [Tenable] Ditto. Tenable’s onboard, too, using OpenAI’s models to create the CyberAgents Exchange AI Inspector within the Exchange (launched August). Exchange Inspector combines frontier assessment using OpenAI GPT cyber models, skills inspection powered by Tenable One AI Exposure and expert review from Tenable researchers, helping to enable security teams to safely accelerate enterprise adoption of agentic AI. The collaboration grew from Tenable’s participation in the OpenAI Daybreak Defense Network. Exchange Inspector is expected to be available in September. Tenable
- [Abnormal AI] Ditto. Abnormal AI’s new AI Cloud Security product uses OpenAI models for detecting risky or malicious AI agents inside customer cloud environments. Abnormal’s behavioral engine builds a living model of every identity in the cloud estate, spanning human identities, service accounts, API keys, and AI agents, to learn what “normal” behavior looks like for each. Like human adversaries, malicious or misbehaving AI agents may create behavioral signals when their activity diverges from those established patterns. Abnormal AI identifies anomalous behavior in cloud environments to surface potential security incidents for investigation. Abnormal AI
- [Doppel] First Canadian office. Doppel opened an office in Toronto – it’s first in Canada; focus is product development and customer growth. The new office will establish a local presence for Doppel as the company increasingly serves Canadian businesses and continues investing in the region’s deep sales and engineering talent. The new Toronto office will primarily serve as an engineering hub, while also providing a local base for members of Doppel’s growing go-to-market team in the area. Doppel is actively building its regional team and plans to grow its Canadian office to 25 employees over the next year. Doppel