AI needs 'human control' to avoid being weaponized, says Microsoft's president

In an exclusive interview with CNBC, Brad Smith, President and Vice-Chairman of Microsoft, emphasized the potential dual nature of artificial intelligence (AI), stating that it can be both a tool and a weapon. He stressed the importance of maintaining human control over AI to prevent untoward consequences and discussed the necessity of having mechanisms in place to slow down or halt AI operations when needed. Smith referenced the risks associated with generative AI-powered technologies like the ChatGPT chatbot, which has raised concerns due to its ability to generate human-like responses. Smith and other tech leaders, including Sam Altman of OpenAI and executives from Google's DeepMind, have expressed fears about the potential risks of AI and called for regulations to ensure its safe development and application. Smith compared the need for safety measures in AI to existing safety protocols in other technologies, emphasizing that AI should be controlled to avoid catastrophic outcomes. He also clarified that AI should serve as a supplementary tool to enhance human capabilities rather than replace human decision-making.

Link to the full article:
ttps://www-cnbc-com.cdn.ampproject.org/c/s/www.cnbc.com/amp/2023/08/29/ai-needs-human-control-to-avoid-being-weaponized-microsofts-brad-smith.html