# AI should be treated like nuclear weapons

https://mins.news/en/video/63627

- DW News, 2026-09-18 03:00 UTC
- Tech
- Original: https://www.youtube.com/watch?v=xVYv7Whs2P8

> AI safety researcher Nate Soares warns that superintelligent AI could become more dangerous than nuclear weapons, with a rogue system potentially threatening humanity. Ahead of Donald Trump’s meeting with Xi Jinping, DW discusses the US-China AI race and whether humans can retain control of the technology they are creating.

## Stories

- [AI Cybersecurity and Loss of Control Risks](https://mins.news/en/topic/ai-cybersecurity-and-control) — Markdown: https://mins.news/en/topic/ai-cybersecurity-and-control.md

## Related coverage

- [US, China security experts propose nuclear-style safeguards for AI risks](https://mins.news/en/item/62078) — Reuters, 2026-09-17
- [Could AI really kill us?](https://mins.news/en/video/61632) — CBC News, 2026-09-16
- [US accepts risk of human extinction if it controls lethal AI weapons, expert says](https://mins.news/en/item/60599) — TASS, 2026-09-16
- [Why top AI researchers are warning about human extinction](https://mins.news/en/item/58952) — DW, 2026-09-15
- [Trump calls AI safety warnings a ‘hoax’ as tech leaders raise alarm](https://mins.news/en/video/58858) — Al Jazeera English, 2026-09-15