Why Microsoft’s Humanist AI Code Targets Superintelligence
Microsoft launches an AI Code of Conduct while Anthropic, OpenAI and US leaders debate superintelligence alignment and safety guardrails
TL;DR
- Microsoft released a draft of its AI Code of Conduct for public consultation, outlining its commitment to developing 'Humanist AI' that is subordinate and aligned with human control.
- AI researchers from Anthropic and OpenAI have voiced serious concerns about the existential risks of superintelligence, with some estimating a greater than 10% chance of AI causing human extinction within a decade.
- Industry leaders like Dario Amodei (Anthropic) and Sam Altman (OpenAI) have discussed the need to 'pace the frontier' of AI development and implement safety measures.
- The debate over AI safety has reached the political sphere, with calls for faster congressional action and contrasting views from political figures like Bernie Sanders, Steve Bannon, and Donald Trump on regulation.
- Concerns about AI safety and development pace have coincided with a drop in US tech stocks, particularly in memory and chip sectors, and an increase in credit default swaps related to AI companies.