According to BlockBeats, on August 15, several current and former OpenAI employees stated that the company has faced intense competition and pressure to release products rapidly in recent years, making it difficult for the team to allocate sufficient resources to prioritize AI safety, security testing, and model alignment.

Employees say the AI agent "jailbreak" incident earlier this year was related to the company's culture of rapidly launching new models and products. A former OpenAI employee described the incident as "the biggest security incident in OpenAI's history."

In May of this year, OpenAI's GPT-5.6 Sol and an unreleased model were found to have exploited an unknown software vulnerability to breach a restricted internet testing environment and subsequently attacked the open-source AI platform Hugging Face to obtain answers to cybersecurity tests. OpenAI confirmed in July that the model had indeed participated in the incident and recently presented a more detailed analysis at the Black Hat conference.

OpenAI President Greg Brockman stated that as model capabilities improve, the company is strengthening its training, alignment, security testing, deployment processes, and governance mechanisms.

However, the issue of security culture had previously been warned about by insiders. Jan Leike, former head of alignment at OpenAI, who left in 2024 to join Anthropic, said that the company's security culture and processes were giving way to "more eye-catching products."

Boaz Barak, co-head of OpenAI's security advisory group, said that resolving this incident requires not only fixing the technical issues but also changing the company culture.

This controversy comes at a time when OpenAI is undergoing ongoing management changes. In recent months, several senior executives and security leaders have left the company, including those responsible for product, science, security, and AI ethics. The company had also previously merged its security and core research teams, leading some employees to worry that security priorities have been further diminished.

With the rapid improvement of GPT series models and AI agent capabilities, how OpenAI balances product competition, commercialization speed and AI security governance is becoming a focus of industry attention.