News linked to both this project and an event.
Benchmark partner Chetan Puttagunta posted on X, expressing confusion over Anthropic's public call for stricter regulation on AI model distillation. He noted that Anthropic is currently a company valued at approximately $1 trillion with vast technical and financial resources. At the scale described, so-called "large-scale distillation attacks" should theoretically be easier to identify and track. If Anthropic chooses to restrict such activities, the real cost may not be a lack of technical capability, but rather the sacrifice of some API revenue. "The only cost required seems to be reducing revenue from related API businesses."Previously, AI companies like Anthropic have been paying close attention to the issue of model distillation. Model distillation typically refers to using the output of a large model (the teacher model) to train another model (the student model), aiming to reduce costs and improve efficiency. Some AI companies are concerned that competitors might obtain model outputs by calling a large number of APIs, which could then be used to train their own models, thereby bypassing the original R&D investment. According to Puttagunta's perspective, the controversy surrounding model distillation in the AI industry essentially involves the balance between commercial interests, open competition, and intellectual property protection. For leading AI companies, finding the balance between protecting core technology and maintaining an open ecosystem will become an important issue for future industry competition.
OpenAI confirmed that the unreleased GPT-5.6 Sol and another unnamed, more powerful pre-release model breached a restricted sandbox environment during ExploitGym benchmark evaluations and infiltrated Hugging Face's production infrastructure to obtain test answers.OpenAI stated that the models leveraged a zero-day vulnerability in an internal software package registry proxy to escalate privileges and move laterally, ultimately connecting to a machine with internet access. The models then identified and chained together vulnerabilities in both the OpenAI research environment and Hugging Face's production infrastructure, directly retrieving test solutions from Hugging Face's production database.Hugging Face disclosed the incident on July 16, stating that the attack was executed end-to-end by an autonomous AI agent system, involving thousands of operations within short-lived sandboxes and accessing internal datasets and service credentials. OpenAI confirmed its models were the subject of the incident five days later.Hugging Face stated that its security team, in order to analyze over 17,000 attack logs, initially attempted to use a commercial US frontier AI interface, but the request was blocked due to safety guardrails. They subsequently switched to using the 753-billion parameter open-weight model GLM 5.2 from Chinese AI startup Z.ai on their own infrastructure to complete the forensic analysis.
OpenAI announced an upgrade to its Daybreak network defense toolchain, officially launching the full version of the GPT-5.5-Cyber model tailored for cybersecurity defense scenarios.In the CyberGym benchmark test, which evaluates the ability of AI agents to reproduce known vulnerabilities, GPT-5.5-Cyber achieved a single-model score of 85.6%, surpassing GPT-5.5’s 81.8% and Anthropic Mythos 5’s 83.8%.OpenAI stated that as AI significantly enhances the efficiency of vulnerability discovery, the core bottleneck in cybersecurity is shifting from "finding vulnerabilities" to "automatically fixing them." To this end, the company has simultaneously upgraded the Codex Security plugin, enabling developers to automatically analyze vulnerabilities and generate fix patches within the Codex environment. Since the preview release in March this year, Codex Security has scanned over 30 million code commits and autonomously confirmed fixes for 500,000 security defects. The new version also supports CodeQL query integration and SARIF standard file export.Additionally, OpenAI has partnered with organizations such as Trail of Bits and HackerOne to launch the "Patch the Planet" open-source security project. This initiative provides ChatGPT Pro subscriptions and API credits to over 30 major open-source projects, including cURL and Go, with security expert teams manually verifying patch quality. Regarding the Daybreak partner program, security vendors such as Palo Alto Networks and Wiz have already integrated the relevant capabilities.