OpenAI WHISTLEBLOWER Reveals What OpenAI Is Really Like!

1. Microsoft bypassed safety protocols in releasing GPT-4.

🥇92 01:25

Microsoft released GPT-4 in India without proper approval, highlighting a breach in safety protocols and governance structures.

• Safety board was supposed to approve GPT-4 deployment.
• Microsoft's actions revealed a lack of adherence to safety measures.
• Concerns arose about maintaining relationships over enforcing safety protocols.

2. Cultural shift at OpenAI post internal conflicts.

🥈88 04:48

After internal conflicts like the Ilia coup, a shift in culture at OpenAI led to a lack of respect for safety personnel.

• Safety personnel faced anger and distrust post-conflicts.
• Perception of dishonesty within the old board affected safety personnel.
• Disbanding of the super alignment team indicated significant internal changes.

3. Predictions suggest AGI by 2027.

🥇94 07:02

Consistent predictions from OpenAI employees, including Daniel Kokotajlo, point towards AGI potentially by 2027.

• Short timelines for AGI reflect advancements in AI technology.
• Publicly available information supports the plausibility of AGI by 2027.
• Expectations are based on observable model capabilities and progress.

4. Sam Altman's potential influence in the AI industry.

🥈89 09:23

Sam Altman's role at OpenAI could position him as one of the most powerful figures in the AI world, raising questions about governance and motives.

• Altman's potential power stems from his position at OpenAI and major AI company partnerships.
• Implications of AGI arrival on power dynamics and governance structures.
• Considerations on the significant influence Altman could wield in the future.