SentinelLABS evaluated OpenAI's native compaction feature in the Responses API for automated binary analysis workflows. Compaction reduced input tokens by approximately 86% with no aggregate score degradation, but domain object modeling quality decreased, indicating that structural reasoning can be flattened during context compression. The research advocates for a context-engineering strategy that separates compacted working memory from durable artifact storage and treats compaction as lossy until validated.
OpenAI
7 posts
Context Engineering | Compaction & Agent Memory for Automated Malware Analysis That AI Extension Helping You Write Emails? It’s Reading Them First Unit 42 identified 18 high-risk browser extensions masquerading as GenAI productivity tools that function as remote access Trojans, infostealers, and spyware. These extensions exploit browser permissions to intercept API keys, exfiltrate DOM content, establish persistent WebSocket C2 channels, and dynamically route traffic via malicious proxy configurations.
Intelligence Center Generative AI enables defenders to rapidly deploy highly adaptive honeypots that simulate complex environments like Linux shells or IoT devices. By leveraging LLMs to generate plausible responses to attacker inputs, organizations can deceive automated AI-driven attacks, shifting the defensive strategy from passive detection to active manipulation and intelligence gathering.
Socket Selected for OpenAI's Cybersecurity Grant Program Socket has been selected for OpenAI's Cybersecurity Grant Program, gaining API credits and access to frontier models via the Trusted Access for Cyber framework. This partnership enhances Socket's ability to detect malicious packages in open-source registries like npm and PyPI in near real-time, countering the increasing speed and automation of supply chain attacks.
Axios Supply Chain Attack Reaches OpenAI macOS Signing Pipeline, Forces Certificate Rotation A supply chain attack involving a compromised version of the Axios library (1.14.1) impacted OpenAI's macOS app signing workflow. The malicious package was executed in a GitHub Actions CI pipeline with access to sensitive code signing certificates, prompting OpenAI to revoke the certificates, rebuild applications, and force user updates, though no downstream compromise or data exfiltration was observed.
Protecting Publishing: The Real Cost of AI Bots AI fetcher bots are severely impacting the publishing industry by scraping proprietary content in real-time to feed AI chatbots, leading to a drastic reduction in referral traffic and revenue. Organizations are advised to implement advanced bot management and monetization strategies rather than relying solely on default blocking to mitigate infrastructure strain and financial losses.
Secure Homegrown AI Agents with CrowdStrike Falcon AIDR and NVIDIA NeMo Guardrails CrowdStrike has announced the integration of Falcon AI Detection and Response (AIDR) with NVIDIA NeMo Guardrails to secure enterprise AI agents against runtime attacks. The solution provides programmable guardrails to prevent prompt injection, data exposure, and unauthorized actions by applying over 75 built-in classification rules to LLM interactions.
