Reuters World News - Iran-US threats, Germany elections and AI at the UN

Episode Date: September 21, 2026

The US and Iran trade threats as Houthi attacks escalate. AI is on the agenda at this week's UN General Assembly. Anthropic sets up a biology lab as it ramps up its AI drug program. And Germany's far-right AfD tops a second election in a month. Podcasts, breaking news, and video analysis from the source. ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Download the Reuters app here.⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Listen to the Morning Bid podcast ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠here⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠. Sign up for the Reuters Econ World newsletter ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠here⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠. Listen to the Reuters Econ World podcast ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠here⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠. Visit the ⁠⁠Thomson Reuters Privacy Statement⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ for information on our privacy and data protection practices. You may also visit megaphone.fm/adchoices to opt out of targeted advertising. Learn more about your ad choices. Visit megaphone.fm/adchoices

Hi, I'm Kim Vinal in Wanganui, New Zealand. It's Monday, September 21st. Today, Iran warns a US attack is imminent as Saudi Arabia grapples with Houthi advances. Open AI's Sam Altman prepares to address the UN with AI fears front and centre at the gathering of world leaders. And another win for the far right in Germany puts Chancellor Mertz in a tight spot. This is Reuters World News, bringing you everything you need to know from the front lines in 10 minutes, seven days a week. Iran is warning against any big new attack by the United States and its allies, saying it has intel that a move is being prepared. Tehran didn't elaborate on anything specific, but the U.S. State Department did issue a security alert over the weekend, saying there was potential for unforeseen escalation in the Middle East.

All that adds to the anxiety in the region, already on edge after drone and missile strikes on Saudi Arabia, with the Iran-aligned Houthis claiming they struck the capital, Riyadh. Our chief correspondent for Saudi Arabia, Timor Azari, explains why that matters. It's been many years since Riyadh was targeted on this scale. So it's definitely an escalation and it's just another sign of what a difficult spot the kingdom here is in

as it tries to sort of decide what to do in Yemen. It's been 11 years and a lot of blood sweat and tears and Yemen is still in conflict and the Houthis are still able to take territory and to threaten Saudi Arabia. Timor says the Houthi attacks have shaken the kingdom. Since 2022, Saudi Arabia has been in talks with the Houthis. Houthis, they reached a ceasefire with the group. So Saudi Arabia seemed pretty secure in the belief

that the Houthis would stick to that path despite the escalation in the region. And we've really seen that turn on its head in the last few months. And now the conversation is basically about, you know, can we ever engage the Houthis in the political process again? And if we can't, then, you know, probably there's only a military solution. If you look at Saudi strategy over the the past sort of decade, it's been predicated on this national transformation strategy where you have huge infrastructure projects and you're really trying to turn this kingdom from a place that exported oil to a place that brings people in and that people visit for tourism. And that depends on having a region that's relatively stable.

And clearly, we're not heading in that direction. The widening conflict in the Middle East is just one of the topics on the agenda as world leaders gather in New York this week for the UN General Assembly. They're also going to tackle the risks of artificial intelligence. With Reuters learning that OpenAI CEO, Sam Altman, will also be there to brief leaders. Our UN correspondent at Nies Larkovsky at the UN headquarters has more. One of the really interesting subjects is artificial intelligence.

Leaders from leading AI labs have warned about the sort of breakneck speed of AI development and Antonio Gutierrez actually has called for coordinated efforts to rein in the speed of AI development for a long time. One important elephant, not in the room, but missing from the room or missing from the conversation perhaps, is President Xi Jinping of China. He is skipping the General Assembly of the United Nations and is opting for one-on-one direct conversation with President Donald Trump in Washington, D.C. on Thursday.

This is inevitably going to create a split-screen moment and is going to highlight concerns about where the locus of political power is, whether it is where the two most powerful nations in the world meet in D.C. Or whether it is here in the building behind me. Let's turn to markets now with AI high on the agenda this week. Here's Mike Dolan from our sister podcast MorningBid. Markets entering the week in reasonably stable format,

but they are looking to the Trump meeting on Thursday, I think, as the event of the week. We already had a meeting between Scott Besson and the Chinese Vice Premier He Leng Feng on Sunday, and that has set out the agenda for that, which will include, of course, discussions about air. and its impact and how to keep tabs on it, keep those guardrails on it. And one of the suggestions from them was that there would be a mutual alert system if things got out of hand.

Both China and the United States still think that the existing structures are okay. They don't want to slow down development. But we'll watch what Trump and Xi have to say on Thursday for that. There will, of course, be more immediate issues around trade and the extension. of the November 10 trade deadline for that truce that's in place. So that will be watched very closely too. And in the backdrop, oil prices have fallen a bit, despite a very noisy weekend in the Middle East and in the Ukraine-Russia war.

Thanks, Mike. Morningbid is available wherever you get your podcasts. The far-right alternative for Germany, or AFD, celebrating another state election win, this time in Mecklenburg-Western Pomerania. The result adds to the party's growing momentum nationwide. And with the far-left linker topping the poll in Berlin, Chancellor Friedrich Mertz's conservatives are being squeezed, and he himself is fighting for political survival,

calling the AFD win a disaster. This is a disaster. Here's Reuters correspondent James McKenzie in Berlin. Voters tend to have moved away from the mainstream, the old mainstream parties. The Christian Democrats and the social democrats, the two parties that have basically divided up the chance he has office between them throughout the history of the Federal Republic since the law. They're moving more and more away from the parties towards the AFT. And also towards the left parties we saw here in Berlin.

and that is a sign for Nets himself for the German mainstream parties but also for parties in the rest of the year where there are also elections issued. Russia's ruling United Russia Party is on course for a decisive win in parliamentary elections there, which the Kremlin has framed as a test of support for the war in Ukraine. Early results show the party winning nearly 58% of the vote, with other parties expected to take seats in Parliament, but who broadly back the government on major issues. Critics say Russia's political system leaves little room for genuine opposition,

with most anti-war candidates excluded from the ballot. Anthropic is taking its AI push beyond the digital realm into drug research. The company is quietly building a wet biology lab in the San Francisco Bay Area, and it says its clodd models could help research, just tackle rare diseases. But as tech reporter Jeffrey Daston explains, the new initiative is a big step at a delicate moment. Anthropics AI advances at the same time

as the company and others in industry are calling for a slowdown because of the danger can be a bit head spinning. And the way that Anthropic makes sense of this, the way it talks about this in at least one key way, is that it believes it has to race to the top. It has to be the number one AI provider for the world to take seriously all of its safety concerns.

So precisely because it's at the frontier, it is able to have everyone else pay attention when it says, I think there should be some policy about how fast we scale the AI, or I think we should do this kind of safety research. And so that is partly why Anthropic is still racing forward. It wants that seat at the table to guide the conversation. Jeffrey says Anthropic is already partnering with some drug makers to provide AI to automate lab and other equipment.

Pharmaceutical companies are liking that. Actually, the CEO Novartis has joined Anthropics board. At the same time, Anthropic is navigating what some sources called a trust problem. there is this concern that as those drug companies use Claude for their own drug programs, as they put their own proprietary data into Claude and talk to Anthropic about what they're doing, that Anthropic will learn from that and then match, maybe outcompete those companies in the drug industry. For more on any of the stories from today, as well as more analysis and news shaping our world, out Reuters.com or the Reuters app. There's a link in the description to download it and start

getting pure news straight from the source. Don't forget to follow us on your favorite podcast player. And if you're listening on a smart speaker, just ask for the latest news from Reuters seven days a week. We'll have an update for you this afternoon.