Google DeepMind’s Tulsee Doshi says AI’s next phase depends on user trust

# Google DeepMind's New Challenge: Making AI Trustworthy Enough for Real Life Google's AI team is grappling with a tricky balance as they roll out more powerful AI tools—they need systems that are helpful and responsive to users, but not so eager to please that they give bad advice or bypass important safety guidelines. The company is now checking AI for subtler problems like whether it's just telling people what they want to hear instead of being honest, and whether AI assistants can safely act independently without causing problems.