Key points:
- OpenAI is a company that makes smart computers.
- Some important people left the company because they were worried about the safety of the smart computers.
- The company made a new group to check if their smart computers are safe and how to make them better.
- They also changed some rules so people can talk more freely about the company. Read from source...

- The article is too short to cover the complex topic of AI safety and governance. It only scratches the surface of the issues and does not provide enough details or context for the reader to understand the implications of OpenAI's decisions.
- The article uses vague terms like "safeguards" and "recommendations" without explaining what they mean or how they are implemented. It also does not mention any specific measures or protocols that OpenAI is following or planning to follow to ensure the safety and security of its AI models.
- The article focuses on the personnel changes and conflicts within OpenAI, rather than the technical challenges and opportunities of developing safe and beneficial AI. It portrays OpenAI as a chaotic and unstable organization, which may undermine its credibility and reputation in the field.
- The article does not provide any evidence or sources to support its claims or assertions. For example, it does not cite any research papers, studies, or interviews that show how OpenAI is addressing the long-term AI threats or aligning its models with human values and goals. It also does not mention any external feedback or criticism that OpenAI has received from other experts, stakeholders, or regulators.
- The article mentions Microsoft's new board committee dedicated to evaluating the safety and security of its AI models, but does not compare or contrast it with OpenAI's approach. It also does not explain how Microsoft is collaborating or competing with OpenAI in the AI research and development domain.