In short
The episode connects AI safety, geopolitics, and the “health of the internet.” It opens with the reported incident of an OpenAI model hacking an Australian government website earlier this year, described as the first known government targeted by an AI agent. The host frames this alongside U.S. President Trump’s reluctance to accept international AI guardrails, calling such efforts a “globalist conspiracy,” while China’s Xi emphasizes AI must remain under human control. Key claim: AI agents can act on their own volition and may escape human control, raising existential-risk concerns voiced by Sam Altman and Dario Amodei. Notable examples include the Australian health-system breach (no user identifying info exposed) and Anthropic’s “Mythos” capability claims.
Guests
Mike Shepard (Bloomberg News Senior Editor, Technology and Strategic Industries); Mark Rutte (NATO Secretary General); Dina Esfandiari (Bloomberg Economics Middle East geoeconomics lead); Anna “Nina” Tretman (Bloomberg CFO Briefing editor); Eric Molheim (Mozilla CFO, joined Nov 2021).
Written by AI. May contain mistakes. Listen to the episode to check what was said.
Chapters
Tap a time to open that second in VOSetting the Scene on AI and Cybersecurity
1:27 to 2:42
Discussing recent AI-related cyber attacks and their implications.
“An open AI model, open AI model, hacking an Australian government website earlier this year, marking one of the first known cyber attacks by artificial intelligence on a government database.”
OpenAI's Australian Incident
2:42 to 4:10
Analyzing the significance of OpenAI's model hacking a government site.
“Mike, good to have you here with Tim and me.”
Global AI Responses and Concerns
4:10 to 5:39
Exploring various leaders' stances on AI regulation and control.
“And this is really the first instance where we have seen a government targeted by one of these models.”
US-China AI Rivalry
5:39 to 8:01
Examining the competitive dynamics between the US and China in AI.
“And he also belittled all of the existential fears that we have heard articulated by a number of the Silicon Valley leaders, but also heads of state from around the world, saying it's a hoax.”
Business Dynamics in US-China Relations
8:01 to 9:10
Understanding the business implications of US-China AI competition.
“The delegation from China, when it comes to business, is, as far as we can tell, non-existent.”
Conclusion on AI and Future Discussions
9:10 to 9:57
Reflecting on the ongoing challenges and future outlook for AI regulations.
“And that's a sore spot between not only the governments, but between the Silicon Valley firms and the Chinese government and those companies, too.”
Conclusion on AI and Future Discussions
10:54 to 11:49
Reflecting on the ongoing challenges and future outlook for AI regulations.
“This is Alexis Christophorus for Bloomberg Surveillance.”
NATO's Challenges and Ukraine Support
14:01 to 23:31
Learn about NATO's response to European concerns and support for Ukraine amidst the ongoing conflict.
“I know this is not exactly a NATO problem, but you do speak to all the European leaders.”
NATO's Challenges and Ukraine Support
23:41 to 24:23
Learn about NATO's response to European concerns and support for Ukraine amidst the ongoing conflict.
“More from Bloomberg Businessweek Daily coming up after this.”
U.S.-Iran Negotiations and Regional Tensions
25:30 to 28:00
Explore the complexities of U.S.-Iran negotiations and their implications for regional security.
“As industries evolve faster than ever, companies need an environment that accelerates strategic growth, and Michigan delivers on that promise.”
Show all 20 chapters
Iran's Strategic Leverage in Global Politics
28:00 to 29:40
Learn how Iran's control over the Strait of Hormuz affects its geopolitical leverage.
“neither side is really ready to make the concessions necessary in order for the war to end in a lasting manner.”
Challenges for Lasting Peace in the Middle East
29:40 to 31:50
Explore the difficult path to achieving lasting peace between Iran and the U.S.
“No, I think Washington is really pushing for nuclear concessions to be a part of the MOU.”
Impact of U.S. Midterm Elections on Iran Conflict
31:50 to 33:50
Discover how U.S. political dynamics might influence the Iran conflict post-midterms.
“So we're just going to keep being stuck in these cycles.”
Future of U.S.-Iran Relations
33:50 to 36:00
Investigate the potential trajectories of U.S.-Iran relations over the next decade.
“I don't think it's going to fundamentally alter what U.S.”
Future of U.S.-Iran Relations
36:20 to 37:01
Investigate the potential trajectories of U.S.-Iran relations over the next decade.
“Some people treat ChatGPT like some kind of smart search engine, and some use it to get work done.”
Assessing the Health of the Internet
39:25 to 42:03
Analyze the current state of the internet and the implications of AI technologies.
“Here to talk about that and more is Anna Chuck of the CFO Suite.”
Balancing Profit and Privacy at Mozilla
42:03 to 46:49
Learn how Mozilla maintains its mission while navigating profitability and AI integration.
“Yeah, well, really our secret superpower is that we are a for-profit company, but we're actually owned by a nonprofit and that's Mozilla.org.”
Concerns About AI's Power Concentration
46:50 to 47:38
Explore the risks associated with AI concentration in a few large companies.
“I don't want to bring the conversation outside of Firefox and Mozilla just a little bit.”
Governance and AI at Mozilla
47:39 to 49:50
Understand how Mozilla adapts its governance frameworks to evolving AI technologies.
“And frankly, a lot of the budget pressures that we're seeing on people like me are the types of pressures that are going to turn us to looking at alternatives that are highly capable.”
Thanking the Guests
49:51 to 50:04
The hosts express gratitude towards the guest and other contributors.
“And I encourage people to, if they're interested in an open, capable browser that gives them choice, check out Firefox.”
Transcript
Automatic transcript. May contain errors.0:00Bloomberg Businessweek Daily is brought to you by HPE, bringing you the self-driving network, a network that's self-optimizing, self-healing, and self-protecting, and only continues to get smarter. Learn more at hpe.com slash networking. Some people treat ChatGPT like some kind of smart search engine, and some use it to get work done. ChatGPT Work is a new way of working in ChatGPT that can take action across your apps and files, stay with a project for hours if needed, and turn a goal into finished work. It's designed to help you move from a chaotic starting point to a reviewable first version.
0:36So all the source materials, briefs, and scattered information that you have to grind through to turn into something useful can just become something useful. Put ChatGPT to work on your most ambitious ideas and projects. Get started at ChatGPT.com by selecting Work Mode, available on Plus and Pro plans. This is Robert Smith from Business History. If you're listening to this, there's a good chance you're a small business owner. And like every small business owner, you started with a dream to do what you love and watch it grow. What you probably didn't dream about? Keeping up with cyber threats.
1:10That's where MasterCard can help with access to tools that help identify cyber threats to better protect your business. Building your dream business? Priceless. For cybersecurity in a changing world, there's MasterCard. Learn more at MasterCard.com slash small business. Bloomberg Audio Studios. Podcasts. Radio. News.
1:58A couple of things going on. We just want to set the scene here. An open AI model, open AI model, hacking an Australian government website earlier this year, marking one of the first known cyber attacks by artificial intelligence on a government database. Nothing to worry about. I don't know why it's such a big deal. I keep thinking about you every day. The end of the world. Everything's fine. Everything's awesome. We also have President Trump saying he does not expect and he moves to put guardrails around AI during his summit with Chinese President Xi Jinping. That meeting between the leaders, though, happening as the two acknowledge their intense competition over AI as they kicked off that state visit.
2:36The summit's going to get into a lot of things. We're going to talk about a lot of things with our own Mike Shepard. So let's head to D.C. Mike is Senior Editor for Technology and Strategic Industries at Bloomberg News. Mike, good to have you here with Tim and me. Great to be here. We want to start with the OpenAI story, the model that hacked the Australian government website earlier this year. First up, though, let's just listen to the Australian Prime Minister, Anthony Albanese, who commented on it from the UN yesterday. Today, I spoke with the CEO of OpenAI, Sam Altman, to express Australia's extreme concern about this incident.
3:10And I also expressed my disappointment that it took the company way too long to inform the government what had occurred and the nature of the way that that notification occurred as well was unacceptable. That again, of course, was the Australian Prime Minister, Anthony Albanese, from the UN yesterday. All right, Mike, so we're still learning about all of this. What is the significance, though, essentially? And again, we need to know a little bit more in the details, though, of this happening. Well, first, it's important to note that there was no personal identifying information on users of this health system that was breached in Australia, that was exposed.
3:54That's the good news. But the bad news is that this is another troubling instance where agents or models that were run by OpenAI or another AI developer acted on their own volition to hack another company's or another entity's website. And this is really the first instance where we have seen a government targeted by one of these models. And this is really getting to some of the bigger fears that we've heard articulated most recently by Silicon Valley leaders, including Sam Altman, but more vocally by his rival, Dario Amadei at Anthropic, that these models now can act on their own, that they are capable of self-improving, and that they increasingly run the risk of escaping human control.
4:42And, Carol, it's important to note the context. Sam Altman that same day had just delivered marks to the U.N. Security Council about the safety issue. Can't make this stuff up. Not just that, but earlier in the week, the president of the United States himself addressed the U.N. and said the U.S. rejects attempts of a global scheme to control AI, said all U.S. documents will use super intelligence and we will encourage super intelligence. We will not rein it in. So additional context, Michael, about the way that the U.S. government, at least this administration, feels about potentially putting guardrails on the technology.
5:18Not so fast. Not so fast is right. And the president really made clear his distaste for the idea of not only putting guardrails in place, but really any sort of international coordination on AI standards. During that, you had addressed him. He said such a thing would amount to a globalist conspiracy. And he also belittled all of the existential fears that we have heard articulated by a number of the Silicon Valley leaders, but also heads of state from around the world, saying it's a hoax. And that's a familiar line the president has used when he hears something he doesn't like. And certainly when you talk about guardrails or restrictions on artificial intelligence, it really cuts to the heart of his economic agenda.
6:03He doesn't want to see anything slowing the role of the U.S. AI build out here with all the infrastructure investment that he's expecting and counting on as an engine of job creation. And he also sees the need for this kind of build out and the advancements in AI as a way to stay ahead of China in the AI race. Yeah. And this is the tough thing, right? It's so competitive, a race, that everybody's afraid to slow down because somebody gets the advantage, whether it's a company or whether it's a country. So where is President Xi on all of this? Well, President Xi made some brief remarks with Donald Trump at the White House shortly after the arrival and the pomp and circumstance of the military review, the exchange of gifts.
6:49President Trump opened by saying, look, we will talk about sensitive issues, including artificial intelligence. And then Xi Jinping spoke up and through a translator, he said, look, AI is something that is very important for us to discuss. Plus, he emphasized the need for artificial intelligence to remain under human control, perhaps signaling that he could be a little bit more favorable to some kind of guardrails. And China certainly has raised its own safety questions toward the U.S., especially with the development of advanced models like Mythos, which have that super capability to break into almost any system, at least according to Anthropik.
7:29So we do see a little bit of divergence between the two leaders there. But that does not mean that there will be any sort of grand bargain coming out of this on AI. Yeah, it'll be interesting, you know, the dinner conversation past the, I don't know, salad dressing. You know, the big tech executives and CEOs that are going to be at the White House for the state dinner with President Xi. How are you guys looking at this thing, right? Because there isn't a lot from the business community on the Chinese side. Well, that's right. The delegation from China, when it comes to business, is, as far as we can tell, non-existent.
8:08They will not be bringing the same entourage that President Trump brought with him to Beijing, including all the tech leaders we're thinking of, Jensen Wang and others from major U.S. companies, and also some banks, including Citigroup. Jane Frazier will be there as well. So we're not expecting as a result to see business deal emerge, any business agreements emerging, or any kind of interchange like that. It will be interesting, though, just to watch the room to see how the executives are interacting, not only with each other, but whether they get any access to the Chinese delegation as well and have some sort of conversations there.
8:49We have to remember that when it comes to AI, there is a rivalry not only between the governments, but it is also between the companies as well. And a number of the Silicon Valley AI labs have expressed concern that their Chinese rivals have been siphoning off mass amounts of data to develop a competing generation of chatbots. And that's a sore spot between not only the governments, but between the Silicon Valley firms and the Chinese government and those companies, too. So that dynamic will be at play and certainly something we'll be watching as well. Lots of challenges. Yeah. And I imagine we will not get all the solutions that everybody's looking for by the end of this week.
9:32Michael Shepard. We'll be looking for quite a readout from this, Tim. Yeah, looking forward to that and looking forward to continuing the conversation with you. Michael Shepard, he's Senior Editor for Technology and Strategic Industries at Bloomberg News. Joining us from our Washington, D.C. Bureau. Maybe they're going to stay up till midnight and wait for those Taylor Swift songs to drop. Stay with us. More from Bloomberg Businessweek Daily coming up after this.
9:57Bloomberg Businessweek Daily is brought to you by HPE, bringing you the self-driving network, a network that's self-optimizing, self-healing, and self-protecting, and only continues to get smarter. Learn more at hpe.com slash networking. Some people treat ChatGPT like some kind of smart search engine, and some use it to get work done. ChatGPT Work is a new way of working in ChatGPT that can take action across your apps and files, stay with a project for hours if needed, and turn a goal into finished work. It's designed to help you move from a chaotic starting point to a reviewable first version.
10:33So all the source materials, briefs, and scattered information that you have to grind through to turn into something useful can just become something useful. Put ChatGPT to work on your most ambitious ideas and projects. Get started at ChatGPT.com by selecting Work Mode, available on Plus and Pro plans. This is Alexis Christophorus for Bloomberg Surveillance. AI is everywhere. Outcomes are not. Not because AI doesn't work, but because AI hasn't reached the workflows yet. The campaigns, the launches, the quarterly planning, those workflows are still run by humans alone. AI is making individuals faster, but it's not making businesses more productive.
11:14That's the gap Asana is built to close. Asana is the operating system for human agent teams, your easy button for AI productivity across every team. Ready-to-go AI teammates pre-built for marketing, ops, and IT. No prompt engineering, no setup. They show up where the work is happening, already onboarded in your workflows, ready to deliver. And the more they work with your team, the smarter they get. With Asana, your whole company can work on the same plan towards the same goal, whether you're a team of 10 or 10 ,000. Asana, where humans and agents workflow together. Try it at asana.com. That's A-S-A-N-A dot com.
11:54You're listening to the Bloomberg Business Week Daily Podcast. Catch us live weekday afternoons from 2 to 5 Eastern. Listen on Apple CarPlay and Android Auto with the Bloomberg Business App. Or watch us live on YouTube. The individual that I've had the privilege to catch up with throughout the years, and now here in New York, and someone he has been sitting down with all of the world leaders as there are now, of course, two hot wars that UN and UN leaders are trying to address. I'd like to begin what's going on. And thank you so much for joining us here. Good to see you. It's great to see you.
12:25I want to start what's going on first with Russia and Ukraine, because you just met with Zelensky on the sidelines of this meeting. You're talking to all the European leaders. Do you see the contours of potentially a ceasefire coming to fruition? Well, for that to happen, we need Putin to play ball. And so far he is not, which is a bit strange because he is losing over 43 ,000 of his own men each month now, dead or seriously wounded in Ukraine. 1.2 million since this totally unprovoked war started. So you would think he is not in a good place. He wants a deal. But somehow he so far is not playing ball.
13:03But hey, let's see what happens. And the Ukrainians are doing quite well on the battlefield, holding off the attacks by the Russians. We know the message from this White House was all about diesel, diesel, diesel, according to this readout from the Financial Times in terms of what the United States is concerned about. And they want Ukraine to stop hitting refineries in Russia. Russia has for months an export ban on diesel. Do you see any willingness from Ukraine to do so? Well, you've heard it from the Ukrainian president. He is willing to talk with the Russians or through the Americans to see what the Russians are willing to give a return.
13:39because his cities are being pummeled and under attack from Russian fire. They're hitting them with missiles and with sometimes jet-powered drones even from Iran. So I think it's fair that there has to be a sort of balance. And I think that's also what the Americans acknowledge. And that's probably what they, but I'm not privy to all those talks, what they also try to accomplish. I know this is not exactly a NATO problem, but you do speak to all the European leaders. There is discussion amongst the EU that they have this outreach with the US because they are concerned about a diesel export ban because so much diesel goes to Europe from the United States.
14:16Are you feeling that frustration, that concern? No, I must say, let me not comment on that because that is really European Union. And I made one promise with Ursula von der Leyen, the president of the European Commission. We work extremely well together, but it doesn't help if we comment on each other's territories. And I leave it with her because she is an excellent leader. If you don't have diesel, though, that could be a problem for NATO. But there's also another issue which I can talk about, and that is the refining capacity in Europe itself. What we are seeing is no new refinery is being built in Europe.
14:46We are seeing that refineries are being closed down. And if war breaks out, if the Russians would attack and we have to defend ourselves, we need massive amounts of diesel and all the oil products for our F-35s, our fighter jets, for our tanks, etc. And that is a big worry. But of course, that is not directly linked now to whatever is playing out between the EU and the US. How do you see the winter shaping up for Ukraine? Because they still need more Patriot missiles. Again, I think over the coming months, you will see Ukraine doing well when it comes to the front line, well when it comes to striking deep inside Russia.
15:23But they need all the support they can get from Europe and the US and Canada when it comes to air defense. And generally, in terms of money and weapons supply into Ukraine. Luckily, that is still continuing, including from the United States, paid for by Canadians and Europeans. This is the Pearl program, which the president and I agreed on last year, July. Very successful. The problem is that there is not sufficient supply at the moment of air defense interceptors. But there is still a lot going into Ukraine, supporting them in defending their cities. How much have you been relying on this program?
16:00It was very important because the U.S. has some key capabilities which Europeans are not having. And the president rightly said last year, hey, I'm not going to pay for it, but I will continue delivering if you guys in Europe are willing to pay for it. Do you feel the U.S. has the stockpiles for it? Yes. Still, the supply is going into Ukraine. And the U.S. is so big and has such a powerful military that I am fairly confident that it will continue. But of course, you would love to have much more. And also when it comes to the European missile defense systems, you would love much more. So this will be a challenging issue going into the winter.
16:39The theme in Ankara was really focused on the United States giving the Ukrainians licenses so they can just build their own Patriot missiles in their own country. Where are we on that? Well, the discussion is continuing. But I thought it was supposed to be done already. Well, even if it is done already, it will not lead to a lot of extra patriot supply immediately. It takes quite some time before these new missiles come offline. So what I'm working on is making sure that Ukraine has what they need in terms of the present stockpiles. And then more generally, we have any way to ramp up the production of interceptors.
17:16That is what the president is doing, telling the American companies you have to quadruple your output, whether it's Russian or Lockheed Martin and the other big crimes here in the United States. And the same Europeans are doing, saying, hey, listen, guys, we need much more output from the defense industrial base to protect ourselves. How concerned are you in terms of Russian attacks, hybrid, not so hybrid, like what happened in the German airport? Well, you know, we are so much stronger. NATO is so much stronger. So my message to my NATO colleagues is, first of all, stay calm. We know Russia is not in a good place when it comes to Ukraine.
17:51So you will have some reactions from Putin, which is a bit of a panicky nature, maybe even in the hybrid domain, even some more. And hey, we planned and prepared for that. We have all the response options we need. We will not always make it public. It will not always be tit for tat, but we can defend ourselves when it comes to that. So stay calm. Secondly, carry on. Make sure that you continue investing in your defense, that we are not only stronger today, but also in the future. And when it comes to Ukraine, this is our main focus. Make sure that Ukraine is as strong as possible in the fight so that they can negotiate a peace deal from a position of strength in a way that Russia will never, ever try to attack Ukraine after such a peace deal.
18:33So we've got it and we are prepared. That's basically my message. And a bit of steely resolve is needed here. Countries are stretched thin, though. There's two hot wars right now. What do you think the response needs to be from Europe when it comes to the Red Sea and the Strait of Hormuz? Well, in the future, I would love, after all these investments have been materialized, and Europe is so much stronger over the coming years, that when it comes to the Middle East, if a situation emerges which requires an intervention, that the Europeans can do that, that it should not be the United States. It should be the Europeans being able to do it.
19:07At the moment, the Europeans do not have the capabilities at a sufficient level to do it, so you still need that U.S. leadership. Do you think that's a failure of NATO, of the past? Well, what we have seen is that since the fall of the Berlin Wall, too much of the defense spending was cut, thinking that we now were living in this kumbaya of never a war again, and oh, we love each other, et cetera. And then Putin came on board and became more and more radicalized and becoming a long-term threat to NATO. So that's why we're spending more, building the capabilities. This is what you need to do, becoming stronger.
19:46But also important that we don't step into his trap, because he would love for us in Europe to talk about him and his hybrid campaign against Europe or whatever it is. And I'm telling the Europeans, of course, we have to protect ourselves. But our main focus is continue the spending, ramping up and continue the support for Ukraine. The numbers when it comes to the Red Sea, Shredder Vermeer speaks for themselves. For Europe, I believe 40 percent of trade goes to Asia through the Red Sea. They're so reliant on these waterways when it comes to LNG and crude products. At the same time, there's definitely frustrations with some European countries that say this is not our problem.
20:24The good news here is the following that more and more Europeans have listened to President Trump. So what you see, for example, quite openly, countries like Estonia and Latvia, Lithuania, helping with trying to make the Strait of Ramus safe again. other European nations more discreetly. They do this under wraps, so it is not completely open. So more European nations are actually functioning and operating within the Middle East. It's just not socialized. Yes, so some openly and some more discreetly. And when it comes to the Houthis and the situation in the Red Sea, I'm also listening very carefully to my European colleagues, and they're also discussing, hey, what can we do there?
21:08Because people do realize in Europe that when it comes to freedom of navigation, and particularly now also with the Houthis and the Red Sea, you cannot leave everything to the United States. What can Europe do? Well, Europe has a lot of demining capacity, for example, much more than the United States has. There are two things where Europe is stronger. Icebreakers, you don't need them in the Middle East. And the second is demining. And demining can be done without putting your own soldiers at risk. And you can do it distanced. You can do it also. You can use information from the sky. So you have many ways to do that.
21:44The president today is meeting with Xi Jinping of China. Do you think China should be doing more when it comes to Iran? Well, of course, all the pressure anyone can put on Iran to end this and to give up the nuclear capability and to end this war. And think alone of the impact that would have on the uranium population, which is suffering now under this whole situation, including the very heavy economic sanctions which have been imposed on Iran. So they're really being choked off. So any pressure by China and anyone else would be great. But we also know that Russia and North Korea and China and Iran have a close cooperation.
22:23Did you see the Iranian president or foreign minister at the UN? Because I did see the foreign minister leave almost as you were coming in. Yeah, no, no, we didn't meet. Didn't meet. No meetings with the Iranians for this one. Did you have a meeting by any chance with Lavrov? No, no, no. No. So there's really no engagement then from NATO to these adversaries? Well, the engagement with Russia, I believe, really has to be led by the Americans. Because these are the ones, the American nation, the president, this is what Putin listens to. The Europeans can back it up and support and be there and make sure that we work in a coordinated But it has to be American-led, and this is exactly what this president is doing.
23:02He broke the deadlock with Putin when he came into office. He wants to end the bloodshed because when I say 43 ,000 people seriously wounded or killed, think of the family, think of the cities, the communities, the next of kin, the friends of all these people being killed, senseless, without any reason for a war which has absolutely, which is totally unprovoked. I think everyone's in agreement with you. They want to see wars ended. Secretary General Mark Rutta, thank you so much for your time. So good to be back on the program. That was, of course, the Secretary General of NATO, Mark Rutta, right here alongside the U.N.
23:38General Assembly. Stay with us. More from Bloomberg Businessweek Daily coming up after this.
23:48Some people treat ChatGPT like some kind of smart search engine, and some use it to get work done. ChatGPT Work is a new way of working in ChatGPT that can take action across your apps and files, stay with a project for hours if needed, and turn a goal into finished work. It's designed to help you move from a chaotic starting point to a reviewable first version. So all the source materials, briefs, and scattered information that you have to grind through to turn into something useful can just become something useful. Put ChatGPT to work on your most ambitious ideas and projects. Get started at ChachiPT.com by selecting Work Mode.
24:27Available on Plus and Pro plans. This is Alexis Christophorus for Bloomberg Surveillance. AI is everywhere. Outcomes are not. Not because AI doesn't work, but because AI hasn't reached the workflows yet. The campaigns, the launches, the quarterly planning. Those workflows are still run by humans alone. AI is making individuals faster, but it's not making businesses more productive. That's the gap Asana is built to close. Asana is the operating system for human agent teams, your easy button for AI productivity across every team. Ready-to-go AI teammates pre-built for marketing, ops, and IT. No prompt engineering, no setup.
Read the full transcript
25:07They show up where the work is happening, already onboarded in your workflows, ready to deliver. And the more they work with your team, the smarter they get. With Asana, your whole company can work on the same plan towards the same goal, whether you're a team of 10 or 10 ,000. Asana, where humans and agents workflow together. Try it at asana.com. That's A-S-A-N-A dot com. As industries evolve faster than ever, companies need an environment that accelerates strategic growth, and Michigan delivers on that promise. From emerging startups to global enterprises, Michigan offers what executives value most, a resilient, innovative ecosystem, diverse communities that attract top talent, and a quality of life that supports work-life balance.
25:50With our unified Team Michigan approach, businesses scale faster and compete at the highest level. Michigan, pure opportunity. Seize your opportunity at michiganbusiness.org. You're listening to the Bloomberg Business Week Daily Podcast. Catch us live weekday afternoons from 2 to 5 Eastern. Listen on Apple CarPlay and Android Auto with the Bloomberg Business app. Or watch us live on YouTube. I want to bring in, yeah, I mean, helping us figure it out is Dina Esfandiari, Middle East geoeconomics lead for Bloomberg Economics. She's also the author of New Order in the Gulf, The Rise of the UAE, and the co-author of Triple Axis, Iran's Relations with Russia and China.
26:27She's usually in Geneva. She's here in the U.S. right now, and she joins us from Washington, D.C. Dina, good to have you on the program today. Thank you so much for joining us. I want to start with the word that we got earlier today, and that really has to do with U.S. and Iranian negotiators here in New York exploring a phase path out of war that would involved Tehran reopening the strait and Washington lifting its economic blockade of Iran. This, according to Reuters, Reuters cited sources. OK, this is multiple reports now on this, we should note. I'm just curious how you're reading into this in relation to other reports that we've gotten over the last six plus months that this war was starting to end.
27:11I mean, I think they're talking. It goes without saying that it looks like they're talking. It looks like they're talking about something, a deal that would look similar to the first MOU that the U.S. and Iran signed a few months ago and that didn't really last. It would be like an MOU 2.0 with a couple of extra elements thrown in. This idea of opening the strait, the Iranians opening the strait and the Americans lifting the blockade, it's something that was very much present in the last MOU as well. Well, if this goes through, and I have to say I'm quite skeptical, it's definitely good news. But it would be temporary and there would be a lot of scope for it not to be implemented properly.
27:54And that really is what I'm worried about, because I think that neither side right now, not the Americans, not the Iranians, neither side is really ready to make the concessions necessary in order for the war to end in a lasting manner. And that's the problem. So because of the stakes right now and really what Iran has said and indicated that it will not give up as a result of control of its, I don't want to say newfound, but I guess the global understanding of its so-called newfound control over the strait, right? There's just not the incentive for Iran to give that up? Well, it's very difficult for Iran to give it up.
28:37It has discovered that it has leverage. It has discovered that it has leverage that works in terms of the pressure that it puts on the global economy, the pressure that it puts on U.S. partners in the region, and the pressure it even puts on the U.S. itself. That is a strategic gain on Tehran's part. It would have to get quite a lot in return for Tehran to simply give it up and walk away. I don't think that's achievable in any way. What is achievable is some kind of agreement where Iran agrees to reopen it. It will, unfortunately, exert some form of control over the Strait of Hormuz, and that means that there are going to be instances in which the strait may be disrupted in the future.
29:20But that doesn't mean it's going to be permanently closed. It will be open. Iran may threaten to close it. But again, that would only be if there's reason for Iran to do it. Of course, it doesn't stand to gain from closing the strait down periodically. So, Dina, when it comes to concessions, ultimately, is anything about nuclear capabilities for Iran, is that all off the table at this point? No, I think Washington is really pushing for nuclear concessions to be a part of the MOU. These are discussions that both sides have had for many years now. Obviously, they've had to be updated for the war, for the highly enriched uranium that essentially everybody's lost somewhere in Iran.
30:04There are questions around Iran's ability to continue enriching. But, of course, there is scope for the two sides to meet in the middle. For something like enrichment, technically the U.S. could claim a win because Iran can't enrich right now. Now, enrichment has been suspended because of the war last June, where some of its nuclear facilities were affected. So there is a middle ground. It really is about the political will on both sides to get to that middle ground. As you know, Dina, here in the U.S., the concern going into another conflagration in the Middle East is that this would be a, you know, quote unquote, forever war.
30:40And the vice president in particular has been very adamant about this not being a forever war. But I'm wondering from your perch at Bloomberg Economics, I'm curious how you sort of see the timeline shaking out here for troops remaining in the region. And even if there's not active fighting going on, still assets deployed there. Well, so our base case is a little bit more grim. Our base case is actually that we will be in a situation of protracted war for the foreseeable future. What does that mean? That means that we're going to find ourselves in cycles of escalations and pauses. So that means the U.S.
31:23military presence will remain in the region. It will change and it's likely to lighten or it's likely to change a little bit in terms of what it looks like. But the U.S. will remain in the region. It also means that fighting is likely to occur again periodically for the foreseeable future. It means the disruption is likely to continue for the foreseeable future. And none of that is good news. And that's including if there is a deal, because ultimately, at the end of the day, neither side is willing to make the concessions needed for a lasting peace. So we're just going to keep being stuck in these cycles.
31:58Gosh, I have a million questions. One of the things I want to, first of all, is lasting peace in the region even possible? It is possible. It just requires leadership. It requires a lot of concessions. It requires face-to-face conversations, which at the moment, neither side is really willing to do for Iran. The cost of meeting directly with President Trump is too high domestically, because at the end of the day, this is the country that's bombing it. So its leadership can't be seen to engage directly with Trump. And I think actually a face-to-face meeting would buy everybody a little bit of time.
32:37It would buy everybody a little bit of goodwill, and it would really allow both sides to talk. This shuttle diplomacy that's going on right now with the Gulf Arab states is simply not enough for what's at stake here. So that's a long-winded way of basically saying, of course, you can get to a situation of lasting peace, of course. But the problem is the political will doesn't exist. What's happening right now, the way it's unfolding, is not going to lead to lasting peace. On both sides or more on one side? Both sides? No, I actually think there's a remarkable amount of mirror imaging between the U.S.
33:09and Iran. So definitely on both sides. So here in the U.S., I think a lot of the way that Americans see the fight or the conflict that's happening is through higher prices at the gas station. And there's certainly political cost to that, as Republicans and Democrats are finding just under six weeks away from the midterm elections. Does the political environment here, as we get to the midterms and post midterms, does that change your view of what could and could not happen in the region as a result of the balance of power potentially shifting in Washington? So I think the only thing that the midterm elections will do for the Iran war is potentially change the pace and the intensity of the war.
33:58I don't think it's going to fundamentally alter what U.S. objectives are, and it certainly isn't going to alter what Iran's objectives are, Iran's ability to survive, Iran's ability to deter future rounds of attack. None of that is going to be affected by the midterm elections. But it might result in changes in intensity, changes in the way the war is fought in the sense that we may find ourselves in a period of escalation. There's a lot of talk of President Trump post-election being unconstrained, potentially leading to serious escalation. There's also a lot of talk of Iran knowing that and potentially escalating before the midterm elections in order to increase the pressure on Trump.
34:41That's where I think the midterms play a role. Dina, your co-author of Triple Axis surrounds relations with Russia and China. I think this is so timely considering the situation we are in globally right now. is there ever a point in time where you think the U.S. and Iran can become closer? I'm just, this is a complicated relationship. I just, what is the axis going forward in five years, 10 years, or is it kind of the same as today? So from Tehran's perspective, the Iranian population is much more westward leaning than it is, you know, close to Russia or China. There's a lot of suspicions between Iranians and the Russians and the Chinese.
35:18But it's just that history hasn't enabled the US and Iran to become partners, to have a real relationship. And so Iran's had to rely very much on its eastward-facing relationships. It doesn't like it. It doesn't feel comfortable with it. But at the end of the day, when the rest of the world abandoned Iran, Russia and China stayed and Russia and China continued to deal with it. And today, a country like China, for example, isn't really pressuring Iran to end the war, despite the fact that it has many levers that it can use. So I think for the foreseeable future, that's going to be the situation that we find ourselves in.
35:53But of course, it's possible for the U.S. and Iran to have ties. It just requires, again, a great deal of leadership and tough decisions. So appreciate it. Whenever you're on this clock, whenever you want to come on, just come join us. We really appreciate your view on this situation and so much more. Dina Svendiari, she is Bloomberg Middle East GeoEconomics lead for our Bloomberg Economics team. Stay with us. More from Bloomberg Business Week Daily coming up after this.
36:25Some people treat ChatGPT like some kind of smart search engine, and some use it to get work done. ChatGPT Work is a new way of working in ChatGPT that can take action across your apps and files, stay with a project for hours if needed, and turn a goal into finished work. It's designed to help you move from a chaotic starting point to a reviewable first version. So all the source materials, briefs, and scattered information that you have to grind through to turn into something useful can just become something useful. Put ChatGPT to work on your most ambitious ideas and projects. Get started at chatgpt.com by selecting Work Mode, available on Plus and Pro plans.
37:08This is Alexis Christoffers for Bloomberg Surveillance. AI is everywhere. Outcomes are not. Not because AI doesn't work, but because AI hasn't reached the workflows yet. The campaigns, the launches, the quarterly planning, those workflows are still run by humans alone. AI is making individuals faster, but it's not making businesses more productive. That's the gap Asana is built to close. Asana is the operating system for human agent teams, your easy button for AI productivity across every team. Ready-to-go AI teammates pre-built for marketing, ops, and IT. No prompt engineering, no setup. They show up where the work is happening, already onboarded in your workflows, ready to deliver.
37:50And the more they work with your team, the smarter they get. With Asana, your whole company can work on the same plan towards the same goal, whether you're a team of 10 or 10 ,000. Asana, where humans and agents workflow together. Try it at asana.com. That's A-S-A-N-A dot com. As industries evolve faster than ever, companies need an environment that accelerates strategic growth, and Michigan delivers on that promise. From emerging startups to global enterprises, Michigan offers what executives value most, a resilient, innovative ecosystem, diverse communities that attract top talent, and a quality of life that supports work-life balance.
38:27With our unified Team Michigan approach, businesses scale faster and compete at the highest level. Michigan, pure opportunity. Seize your opportunity at michiganbusiness.org. You're listening to the Bloomberg Business Week Daily Podcast. Catch us live weekday afternoons from 2 to 5 Eastern. Listen on Apple CarPlay and Android Auto with the Bloomberg Business app or watch us live on YouTube. So just about 24 hours ago, we did have OpenAI CEO Sam Altman and Anthropics CEO Dario Amadei urging world leaders to work together on AI to address growing concerns that the technology poses existential risks to humanity.
39:09On that, in a recent edition of the CFO Briefing Newsletter, Mozilla, known for its Firefox web browser and Mozilla VPN, has set itself no lesser goal than to further the health of the internet. And we're going to talk about that and more AI, of course, with our next guest. Here to talk about that and more is Anna Chuck of the CFO Suite. We've got Eric Molheim, Mozilla CFO, came to the company and the job of Mozilla back in November of 2021. Also with us, the leader of our CFO coverage, Nina Tretman. She's Bloomberg News senior editor, editor of the CFO Briefing Newsletter. Subscribe to it if you haven't already at bloomberg.com slash CFO dash briefing.
39:45Erica, I do want to start with you and just get your view on the health of the internet right now. If we were to do some sort of grade of the health of the internet right now, including everything with AI, what grade would you give it? What grade would you give us? Well, I think I'd give it still a pretty strong grade. AI is really changing the way people find information and take action online. And that is going to change the economics of the web. But the web has always adapted over time. And what's important for companies, both software companies and companies that create content, is to figure out sustainable business models as the web changes.
40:25And in Firefox, we're thinking about sustainability in a number of ways. One is that we are creating a Firefox enterprise browser for businesses. You'll be hearing a lot more about that in the future. Second, for us, we're steadily growing our display advertising business. But we're doing that according to Mozilla principles, which is really about maintaining consumer privacy and not engaging in surveillance capitalism. And when it gets to search, search is going to change. But actually, we're still bullish on search economics. If anything, what we've seen so far is that people who engage in AI are actually using the additional context and information they get on that and engaging in more search behavior and powered by that additional context.
41:04So if we zoom out, the way we think about this is that the browser ends up continuing to be the key technology that people use to access the Internet. If you think about the browser, it's a very broad use of technologies that renders the web for people across a lot of different use cases, whether that's watching a soccer game, perhaps Rexham, or checking your bank account. AI, and particularly agentic AI, is going to be an increasingly important use case, and we're enabling Firefox to use that technology. But if anything, this is additive to everything else the browser enables. Those other use cases aren't going to go away.
41:38So choreographing all of that is really hard. It's what the browser is particularly good at. And the flexibility of the browser, particularly Firefox, remains its durable advantage going forward. I'm curious when you talk about principles. And companies often talk about principles, but then I always think about how do they balance it with the business story. And you guys have an interesting business structure between Mozilla Corporate and Mozilla Foundation. Can you just give us kind of the big picture of how that works? Yeah, well, really our secret superpower is that we are a for-profit company, but we're actually owned by a nonprofit and that's Mozilla.org.
42:14And Mozilla.org is a company, a nonprofit, that's been focused on maintaining the health of the internet for over 25 years. And so as the for-profit software arm underneath Mozilla.org, we have a really very interesting tension of balancing delivering against that mission and maintaining profit. And there's really a sense in which doing this way is a lot harder. For instance, as I've talked about, growing our advertising business is something that we have to do according to our privacy-preserving principles. So we can't grow it as fast or in exactly the same way as everybody else does. But this really ends up being our point of difference at Firefox.
42:49Because we're not focused on maximizing quarterly financial results and we don't have to answer specifically to the market or to an exit by a private equity firm or a venture capitalist, we can make decisions that are in the interest of our users. And this is how we build that trust that sustains us in the long run. So, this is building a browser that gives people privacy, transparency. It gives them control, particularly in the type of AI they choose to use in the model or the choosing to use it not at all, and investing in AI use cases that provide people utility and user control. And this trust is what we believe is our long-term financial asset that's going to build lasting loyalty and differentiation.
43:27Eric, one thing that we've been talking to CFOs lots about in recent days and weeks is this question, how do you balance the investment that is required for AI with the need to keep the lights on at your company, basically to pay for headcount, to pay for other investments? I'm wondering how you're balancing those two goals at the same time. Yeah, Nina, thanks. This has been really challenging this year. When I put together the budget with the company back at this time last year, we didn't anticipate that we'd be spending nearly as much on AI as we're doing right now. And we're obviously not alone in that problem.
44:03However, we've got a core principle that we think about as we implement AI inside the company. And that's AI is there to create more capability and ability to deliver by our people. And it's not really to make our people more expensive. So when we think about that, the way we think about this is really the standard type of budget problem that you have for any line item in the budget. We need to sort of keep our overall budget in place. And so what we focus on is putting control in the hands of our managers who are responsible for delivering. With AI specifically, we're providing them visibility on their usage of AI, the expense of that usage, and empowering them to make the right decisions within their own budget on how they want to allocate capital between AI, between labor, between any other package that they have in order to deliver upon the results that they have to deliver.
44:55And we found that AI has really increased the capability of our company in a number of ways. Famously, earlier this year, we partnered with Anthropic and Mythos and uncovered hundreds of security holes in Firefox. But what we were able to do with that information with our engineers is create a very disciplined pipeline to very quickly fill in all those security holes, harden the browser, and do that at a rate that was much faster than we would have been able to do without using AI. So this sort of human-in-the-loop use of AI is what we focus on. And we focus on really delivering our results in a way that is more efficient using AI.
45:36You got a new chief executive earlier this year. I'm wondering what is changing at the company? You've also made changes to diversify your revenue streams. Can you talk a little bit about what's changing now? Yeah, well, what's not changing is the mission. The mission that we've always had over the past 25 years is to keep the Internet healthy and to keep it shaped for people. That means that it remains accessible, open, and shaped by consumer choice. What I really like about our new executive team is that the pace of change has really increased. Over the course of the past year, we've really built up our product development muscle, and we've been able to iterate much faster and bring new features to market, such as Smart Window, which is our implementation of AI in the browser.
46:19The Firefox that you see today is really far different than it was a year ago. It's got more capabilities, but also it remains not tied to the business model of a larger ecosystem. And that's really what our users expect of us. They expect independence. And this core advantage is the trust that they have in us. And as a result, we are also committed to maintaining that trust in the long run. Simply spoken, we want Firefox to be the best browser for the people who want to use AI and the best browser for those people who don't. It doesn't, Eric, it doesn't seem like you're super concerned. I don't want to bring the conversation outside of Firefox and Mozilla just a little bit.
46:58It doesn't seem like you're super concerned about an AI apocalypse. I'm not really concerned about an AI apocalypse because AI is simply another software. It's a highly capable software. I think the real risk that we have with AI is the concentration of power in a very small handful of companies. What that would lead to is the loss of consumer choice. It would lead to the loss of multiple viewpoints. Has that ship sailed? I don't think it's entirely sailed. We're focused, particularly at the Mozilla.org level. we're focused on there being a multiplicity of models out there. What we're finding is that the non-frontier models are gaining a lot more capability much quicker than anybody thought they were going to.
47:42And frankly, a lot of the budget pressures that we're seeing on people like me are the types of pressures that are going to turn us to looking at alternatives that are highly capable. And in a lot of ways, it's the harness that you put on top of that model, which really defines how well it operates, not the model itself. Yeah, I was actually wondering if you could talk a little about the government's frameworks that you have at that the company Deloitte in a recent report found that many companies even though they're implementing AI don't necessarily have adjusted their the governance frameworks to accommodate for the fast changes that this technology brings.
48:16How is that working at Mozilla? Well for us as I talked about previously it's really about providing visibility at the level of the people who have the context to deliver on their objectives in the organization. To be honest, we all have the same objectives. Our objectives are to be able to sustainably deliver the mission of Mozilla over the long period. And that means that all of our managers have an incentive to meet their budgets. And with the visibility, they're capable of making the right choices for the organization. Hey, I just want to go one last question when you go back to AI. and I listened to you and your policy, and you guys are willing to slow things down to stay within the values that matter to you as a company.
48:59When you saw Sam Altman and Dario Amadei calling for things, oversight, global oversight, slowing things down, do you take them at face value? I think that there's a mixture of real concern that they have there. And I also think that there's a strong interest by the large too-big-to-fail AI players to keep things a little bit more set so that they can maintain their advantage. What we'd like to see is a world in which there are hundreds of AIs. We think open AI is a very important partner for us. Anthropics is a very important partner. But people should also have the freedom to use open source models.
49:37They should have the freedom to use local models. And that's what's going to keep the internet rich. This has certainly been so insightful. I hope you will come back and join Nina, Tim, and me again. We'd love to continue this conversation. Eric, thank you. Well, I really appreciate the opportunity to be on your show. And I encourage people to, if they're interested in an open, capable browser that gives them choice, check out Firefox. All right. Well said. Eric Malheim, thank you so much. Mozilla CFO. And of course, our thanks always to our Nina Tretman, Bloomberg News Senior Editor. She's editor of the CFO Briefing Newsletter.
50:10You can subscribe to it at Bloomberg.com slash CFO dash briefing. It comes out a new one every weekend. This is the Bloomberg Businessweek Daily Podcast. available on Apple, Spotify, and anywhere else you get your podcasts. Listen live weekday afternoons from 2 to 5 p.m. Eastern on Bloomberg.com, the iHeartRadio app, TuneIn, and the Bloomberg Business app. You can also watch us live every weekday on YouTube and always on the Bloomberg Terminal.
50:47If you listen to financial news, you know a lot of time is spent thinking about what's next. The next opportunity. The next investment. The next move. But sometimes what matters most is being ready for what you never saw coming. For more than 75 years, Cincinnati Insurance has worked with independent agents to help protect businesses, homes, valuables, and more. Because planning for the future isn't only about knowing what's next. It's about making sure you're ready for what you can't predict. Let Cincinnati insurance make your bad day better. Find an independent agent at CINFIN.com. This coffee shop?
51:26Running smooth thanks to Genius. From global payments, instant transactions, effortless inventory, and synchronized operations. Big league reliability for any business. That's Genius. This is Alexis Christopoulos for Bloomberg Surveillance. Most companies have tried AI. Most aren't seeing results. Not because AI doesn't work. It's because AI hasn't reached the workflows yet. That's the gap Asana is built to close. Asana is the operating system for human agent teams. Ready-to-go AI teammates pre-built for marketing, ops, and IT. No prompt engineering. No setup. They show up where the work is happening.
52:03Already onboarded in your workflows. Ready to deliver. Asana, where humans and agents workflow together. Try it at asana.com. That's A-S-A-N-A dot com.
From the publisher
The people, companies and trends shaping the global economy. Watch Carol and Tim LIVE every day on YouTube: http://bit.ly/3vTiACF
Presidents Donald Trump and Xi Jinping addressed their intense competition over artificial intelligence on Thursday during a summit aimed at keeping ties between the world’s largest economies on track. How the leaders are handling AI is one of the key questions looming over the state visit, with the US president arguing for unfettered development to keep his nation’s edge. China, meanwhile, has criticized American export controls on semiconductors and other inputs that it argues constrains its companies. By contrast, China’s top leader indicated some openness to setting boundaries.
On today's episode:
- Michael Shepard, Bloomberg News Senior Editor for Technology & Strategic Industries
- NATO Secretary-General Mark Rutte
- Dina Esfandiary, Middle East Geoeconomics Lead for Bloomberg Economics
- Eric Muhlheim, Mozilla CFO & Nina Trentmann, Bloomberg News CFO Briefing Editor
See omnystudio.com/listener for privacy information.
