OpenAI CEO Sam Altman has recently addressed concerns surrounding a controversial clause in the company’s exit agreements that suggested the potential for equity cancellation amid recent resignations at OpenAI. Altman reassured that OpenAI has never utilized this clause and guaranteed that vested equity will not be affected by separation agreements or non-disparagement agreements.

Altman admitted that the clause in previous exit documents was a mistake and should not have been included, taking full responsibility for the error. He mentioned that the company has been updating its standard exit paperwork to avoid such issues in the future.

Concerns arose regarding the equity cancellation provision and its potential misuse, prompting Altman to acknowledge the mistake and assure employees that the company is making changes to prevent such problems moving forward. The resignations of key figures like Jan Leike and Ilya Sutskever have brought attention to OpenAI’s internal strategies and priorities, with critics accusing the organization of focusing more on product development than AI safety.

Despite the restructuring and departures, OpenAI remains committed to AI safety, with Altman and President Greg Brockman emphasizing the importance of ongoing safety research. They aim to raise awareness about the risks and opportunities of AGI and advocate for international governance in the field of AI. Brockman expressed gratitude to departing employees and confirmed the company’s dedication to addressing safety concerns rigorously.

The path to safely developing and deploying AGI poses complex challenges, requiring continuous improvement in safety measures and oversight according to Altman and Brockman. The company is determined to prioritize AI safety and address concerns related to advanced AI systems.