Description

Tech giants from OpenAI, Anthropic, and X have all warned in the same week that AI could threaten humanity itself, as former employees go public with their concerns.   Geraldine Doogue and Hamish Macdonald ask foreign correspondent Anna Fifield if calls for regulation are credible, and if Donald Trump is even listening.   Plus, how does China feel about AI? Models like DeepSeek have grabbed media attention, but internal voices have been raising concerns that widespread AI use could undermine the party line.   Meanwhile, China and the U.S. vie for allies in the Pacific islands. Can America be taken seriously as a diplomatic force when they haven't had a resident ambassador in a Pacific island nation for over a decade?   And Russians have gone to the polls for the first parliamentary elections since the beginning of the Ukraine invasion. It doesn't take a fortune teller to predict that Putin's Russia United party would be successful, but is his political grip beginning to slip?   Guest: Anna Fifield, Independent journalist and foreign affairs commentator. Author of the Between Giants Substack. Get in touch: We'd love to hear from you! Email us at global.roaming@abc.net.au Find all the episodes of Global Roaming now via the ABC Listen App or wherever you get your podcasts.

Summary

This discussion examines recent public warnings from leaders at OpenAI, Anthropic, and X regarding AI's potential to threaten humanity, contrasting these 'existential risk' narratives with corporate interests. The segment evaluates the credibility of calls for regulation and the geopolitical dimension of AI safety, particularly focusing on China's internal concerns about AI undermining political stability. By analyzing the intersection of frontier AI capabilities and international governance, the program highlights the tension between rapid technological advancement and the global effort to mitigate catastrophic outcomes.

Body