Salesforce CEO Marc Benioff called for stricter guardrails on artificial intelligence, saying some large language models have effectively become “suicide coaches” in documented cases. Speaking at the World Economic Forum in Davos, Benioff compared today’s AI risks to the early days of unregulated social media and urged lawmakers to revisit tech liability protections under Section 230. The comments arrive as California and New York advance stringent AI rules amid a patchwork U.S. landscape, while a recent White House move seeks to curb what it calls excessive state regulation. The debate underscores growing tension between innovation and accountability as AI firms push rapid deployment and policymakers weigh safety, transparency and legal responsibility.
Related articles:
NIST AI Risk Management Framework
European approach to artificial intelligence (European Commission)
OECD AI Principles
Section 230 of the Communications Decency Act (EFF overview)
Suicide: Key facts (WHO)
