Mustafa Suleyman, CEO of Microsoft AI, recently emphasized the urgent need for effective AI safety measures and regulation. He criticized the current state of AI alignment, suggesting that while progress has been made, significant risks remain, particularly highlighted by recent incidents involving AI models. Suleyman's comments come alongside the release of Microsoft's 'Humanist AI Code of Conduct,' which outlines the company's principles for responsible AI development.
Suleyman expressed concerns about the philosophy of companies like Anthropic regarding AI consciousness, arguing that it complicates the alignment debate. He pointed out that while AI models have become more capable of following complex instructions, the potential for unintended consequences underscores the necessity for stringent containment measures to ensure safety and ethical use of AI technologies.