Ask Onix
Call for urgent AI threat research
Google DeepMind's CEO, Sir Demis Hassabis, has stressed the need for immediate research into artificial intelligence risks, warning of serious threats posed by the technology.
Global summit highlights regulatory divide
At the AI Impact Summit in Delhi, tech leaders and politicians pushed for stronger global governance of AI, with many advocating for coordinated regulation. However, the U.S. rejected this approach, arguing that centralized control would hinder progress.
White House technology adviser Michael Kratsios stated, "AI adoption cannot lead to a brighter future if it is subject to bureaucracies and centralized control."
Key risks identified by industry leaders
Sir Demis highlighted two primary threats: the misuse of AI by malicious actors and the potential loss of control over increasingly powerful autonomous systems. He emphasized the need for "robust guardrails" to mitigate these risks.
When asked if Google DeepMind could slow AI development to allow more time for safety research, he acknowledged his company's role but noted it was "only one player in the ecosystem."
U.S. and global leaders clash over AI governance
OpenAI CEO Sam Altman and Indian Prime Minister Narendra Modi echoed calls for urgent regulation, while U.S. officials firmly opposed global governance. Michael Kratsios, head of the U.S. delegation, reiterated the Trump administration's stance: "We totally reject global governance of AI."
UK Deputy Prime Minister David Lammy emphasized collaboration between governments and tech firms, stating, "Security and safety must come first and must benefit the wider public."
Race for AI dominance intensifies
Sir Demis suggested the U.S. and Western nations hold a slight lead over China in AI development but warned that the gap could close within "a matter of months." He stressed the importance of balancing innovation with responsibility, acknowledging that mistakes would occur but asserting that Google DeepMind aims to "get it more correct than most."
Future of AI and education
Looking ahead, Sir Demis predicted AI would become a "superpower" in the next decade, transforming what people can create. Despite advancements, he emphasized the continued value of STEM (science, technology, engineering, and maths) education, suggesting that technical skills would remain advantageous.
He also noted that AI's ability to write code could democratize software development, shifting the focus toward "taste, creativity, and judgment."
Summit concludes with joint statement expected
The AI Impact Summit, the largest global gathering of world leaders and tech executives, wraps up Friday. Delegates from over 100 countries are expected to release a shared statement on AI governance.
