How Block is building AI-powered security with OpenAI technology

Block, the company behind Square, Cash App, and Afterpay, is utilizing OpenAI technology such as the OpenAI Defense Factory and OpenAI Daybreak to scale its security operations. According to James Nettesheim, Chief Information Security Officer at Block, the integration allows the security team to handle vulnerability discovery, detection, and response at speeds unattainable by human teams alone. Work that previously took weeks can now be completed in hours. The security harness routes tasks to different models depending on the specific job, incorporating human validation and oversight for every result. Block has reported a 50 percent boost in triage throughput, an increase in intelligence gathering and investigations, and a new autonomous hunting capability performing roughly 1,100 analyst-hours of investigation per quarter. In recent red team exercises, the model chained together a full attack path in about two hours and moved from a high-level vulnerability to a proposed patch in 18 minutes. The system has also returned roughly 600 analyst-hours per quarter from false-positive triage back to the team for deeper investigations. Nettesheim emphasizes that humans provide vital organizational context and strategy, with the team slowly scaling up the risk level for AI use in sensitive systems as judgment is proven in practice.