2 hrs ago
AI Leaders Urge Caution as China Flags Political Security Threats
Some AI experts are asking companies to slow down their work.
Geoffrey Hinton said people do not know whether very powerful AI could be kept under control.
He said experts are making risk estimates based largely on their personal judgments.
Demis Hassabis agreed that the general direction of slowing down is right, although he said some details need improvement.
China’s intelligence chief warned that AI could be used to attack political and computer systems.
He also said AI might help create malware.
Donald Trump was less worried about these dangers.
He said the United States should stay ahead of China in AI.
The disagreement is about whether AI development should slow down or continue quickly with safeguards.
Geoffrey Hinton endorsed Dario Amodei’s call for AI companies to slow development.
Hinton said experts do not know whether future superintelligent AI can be controlled.
Demis Hassabis said Amodei’s details require refinement but agreed with the overall direction.
China’s State Security Minister Chen Yixin warned that abusive or weaponized AI could threaten national security.
Donald Trump downplayed the concerns, saying the United States should maintain its lead over China in AI.
- Who
- Geoffrey Hinton, Dario Amodei, Demis Hassabis, Chen Yixin, and Donald Trump are the main figures discussed.
- What
- AI experts and officials debated whether artificial intelligence development should slow because of control and security risks.
- Where
- The comments came from Australia’s ABC Radio National Breakfast, China Cyberspace magazine, and Trump’s golf club in Doonbeg, Ireland.
- When
- The warning from Chen Yixin was reported in an article posted Sunday; the other comments were reported in the same coverage.
- Why
- Supporters of a slowdown cited the possibility that advanced AI could become uncontrollable, while Chen warned about political, institutional, ideological, and cybersecurity threats.
AI Safety Advocates
AI Development Advocates
Whether development should slow
AI Safety Advocates
Geoffrey Hinton supported Dario Amodei’s call for AI companies to slow development until the control problem is solved.
AI Development Advocates
Donald Trump argued that the United States should maintain its AI lead and said guardrails could be used to address risks.
How serious the risks are
AI Safety Advocates
Hinton said superintelligent AI could arrive within a decade and that the probability of losing control is uncertain but greater than one percent.
AI Development Advocates
Trump downplayed the concerns, saying some negative claims about AI threats would not happen.
Political and security dangers
AI Safety Advocates
Chen Yixin warned that abusive or weaponized AI could threaten political, institutional, ideological, and cybersecurity systems.
AI Development Advocates
The article presents no separate rebuttal from Chen’s critics, but Trump’s comments emphasize competition and safeguards rather than slowing development.
Key facts
- Main safety concern
- Hinton said nobody knows whether superintelligent AI can be kept under control.
- Hinton’s position
- He called Amodei’s proposal to slow AI development very sensible.
- Hassabis’s position
- He said Amodei’s details need refining, but that the overall direction is correct.
- China’s warning
- Chen Yixin said abusive AI use could threaten China’s political, institutional, and ideological security.
- Cybersecurity risk
- Chen cited AI’s ability to find system flaws and create malware or software exploiting them.
- Trump’s position
- Trump said the United States should remain ahead of China in AI and suggested safeguards could address concerns.
Quotes
Geoffrey Hinton
British-Canadian cognitive psychologist and computer scientist known as the “godfather of AI”
“Nobody knows how to estimate the probabilities of these things. [An Anthropic employee] said 10 per cent. Nobody really knows what figure to give. It's not 1 per cent, it's more than that, and it's not 99 per cent, it's less than that.”
wionews.com
“And we can put guardrails, and we can do this and that, but I think you have a lot of negative forces that are bringing it up and they’re bringing up things that won’t happen,”
wionews.com
Sources
Related news
China Warns AI Could Threaten Political Stability and Security
Trump Rejects AI Slowdown, Says Whoever Wins AI Wins
Trump Rejects AI Slowdown as US Races Ahead of China
Altman and Musk Back Amodei’s Call to Slow AI
China Condemns Calls to Slow Frontier AI Development
Trump Rejects AI Slowdown Calls as Leaders Warn of Risks
AI Whistleblower Urges Self-Regulation as Lawmakers Debate Safety
Nadella Joins AI Safety Debate Over Human Control and Pacing
Anthropic CEO Urges US-China Pact Against AI Bioweapons
Ex-Anthropic Researcher Warns AI Kill Switches May Fail
