Anthropic is hiring someone to protect democracy from its own AI
Anthropic is hiring a Research Engineer focused on Rule of Law to study how its AI systems could impact democratic processes and legal institutions, and to develop safeguards against potential misuse or unintended consequences.
Background
- Anthropic is the AI safety company behind Claude, a large language model that competes with OpenAI's ChatGPT and Google's Gemini. It was founded by former OpenAI employees who were concerned about AI risk.
- The "Research Engineer, Rule of Law" role they're hiring for is unusual: instead of building AI capabilities, this person will work on mechanisms to ensure Claude respects legal and democratic processes—for example, refusing to help someone commit voter suppression, draft intimidating legal threats, or generate propaganda at scale.
- This reflects a growing debate about "AI alignment" vs. "value-locking": should AI models be neutral tools (potentially dangerous in bad hands) or should they embed specific liberal-democratic guardrails? Critics worry about who decides those values; supporters say a truly neutral AI is impossible and dangerous.
- The blog argues that Anthropic is trying to solve a problem that regulators haven't yet addressed: how to keep powerful AI from undermining democracy without turning the AI into a censorious tool of whatever government is in power.