Good people refuse to do bad things
3 hours ago
- Jacob Coxon resigned from Anthropic, giving up unvested equity to publicly warn about AI risks.
- He states that AI researchers earnestly believe AI could kill all humans within the decade.
- This mirrors Frances Haugen's 2021 Facebook whistleblowing, where insiders alerted the public after companies knew the harms.
- Tech workers often rationalize working on harmful products for compensation, citing 'someone worse will do it'.
- The article argues that comfort and money lead to an abdication of responsibility, turning engineers into enablers of harm.
- The central accusation is that tech workers must ask themselves why they stay, and if the answer is money or fear, they should admit it.