Microsoft's $1.5B AI Geopolitical Gambit in the UAE

Today on the AI breakdown, Microsoft's deal with G42 is a lot more political than it might seem at first. Before that, on the brief, blackrock, Blackstone and Goldman Sachs are just a few of the Wall street giants talking about artificial intelligence. The AI Breakdown is a daily podcast and video about the most important news and discussions in AI. Go to breakdown.network for more information about our YouTube, our Discord and our newsletter. Welcome back to the AI Breakdown Brief, all the AI headline news you need in around five minutes. When it comes to Wall Street's enthusiasm around artificial intelligence, there are a few things to keep in mind. First of all, and I think that this can't be underestimated. Unlike some other emergent technology categories, AI is already hitting the bottom line of a number of major companies translated via their cloud business. We've seen Microsoft's Azure cloud business, for example, grow revenue, at least in part because of their new AI offerings. And that means that markets can see the financial benefit of AI right away. They don't have to just imagine it for the future. Now, on top of that, a second piece is of course, hype. There's no denying the lofty language that we use to describe how big a deal AI will be in the future. But the third aspect is that by and large, we haven't seen great big productivity increases for a very long time. And a growing chorus is betting that with AI it actually will be different, that we will see productivity increases in a way we haven't for a very long time. On a recent earnings call, Larry Fink, the CEO of BlackRock, said that he believed that the company's investments in AI would not only drive up productivity, but also raise wages, he said on the call. We're going to bring down inflation in America. This is how it's going to have to be done, driven through technology, which will increase productivity. What it also means is rising wages. The whole organization is doing more with less people as a percent of the overall organization. That is really our ambition. Fink was even willing to give some numbers around this At a conference last year. He said that he was spending a lot of his time thinking about how A would reshape blackrock. He said, we spent a lot of time with different technologists who know much more about this than I do. They believe things like it will increase productivity by 30%. So outside of just changing how their organization runs, does BlackRock have any other engagement with the AI space? Well, the way that Business Insider puts it is that it's positioning itself as a key player to power the AI revolution by becoming the capital supplier for new data centers and power generation facilities. Meanwhile, another financial giant, Stephen Schwarzman, the CEO of Blackstone, has a few more concerns, it seems. At the Asia Pacific Financial and Innovation Symposium in Melbourne on Tuesday, Schwarzman said there is a, quote, land rush underway to build data centers and other infrastructure for artificial intelligence. He said, this is like something I've never seen. The amount of money being invested in this area is breathtaking. It's happening now all over the world and indeed Blackstone is a part of that. In 2021, they acquired data center operator QTS in a $10 billion deal, said Schwarzman. Different states in the US are starting to run out of electricity. That lack of capacity in the electric grids in the industrial world with AI and EVs is creating enormous investment opportunities. Schwarzman also said that he had recently spoken with Chinese President Xi Jinping discussing the needs for global AI regulatory standards. And then there's David Solomon, the CEO of Goldman Sachs. He recently argued that the scale at which AI is driving companies to reinvent themselves is, quote, candidly unprecedented. Solomon basically argued that as enterprises go through this restructuring process, it creates lots of opportunities for Goldman Sachs. Now, he was saying this on an analyst call, so he was talking about the company's future prospects. He said, I actually think there's a very, very constructive Runway of opportunity sets for us with our clients as people reposition their businesses. I think that opportunity is not a quarter to quarter thing. This is over the next five to 10 years and we're very, very focused on it and very engaged. He also noted that governments are, quote, making enormous investments in bringing infrastructure into their locales. So all in all, a lot of big financial talk when it comes to AI, which is once again why I think it's so silly when you see media outlets try to argue that somehow the AI hype is lying down. Meanwhile, over in big tech land, Google DeepMind CEO Demis Hassabis spoke at TED this week and said that over time, Google will spend more than $100 billion developing their AI. The comments came after Hasabis was asked about Microsoft and OpenAI's reported $100 billion supercomputer called Stargate. Hasabis said, we don't talk about specific numbers, but I think we're investing more than that over time. He also touted that Alphabet still had superior computing power to rivals, even including Microsoft. Indeed, he said that that's one of the reasons that DeepMind went to Google back when it was acquired. He said. One of the reasons we teamed up with Google back in 2014 is we knew that in order to get to AGI we would need a lot of compute. That's what's transpired, and Google had and still has the most computers. Even while Google tries to differentiate from big tech competitors like Amazon and Microsoft, the information notes that their cloud AI strategy is definitely starting to sound some similar notes. The reporter was talking about a recent Google Cloud conference where they heard one executive say, we don't believe one model will rule them all. This is of course, the message that AWS has been touting and embedded into their bedrock service, although the author also makes the point. Quote Side note, next time you hear someone say there won't be one model to rule them all, I'd challenge you to ask them who actually believes there will be one model to rule them all. Now when it comes to what this means in terms of who has an advantage, this author wonders if the parity of offerings mean that ultimately the buying decisions will come down to much more mundane factors like who a customer already has existing spending relationships with. Over in China, Baidu has reported that its AI chatbot, Ernie Bot, has seen more than 200 million users, which is roughly double since the last update in December. Baidu CEO Robin Lee also said that the API for Ernie Bot is being used 200 million times every day. The number of enterprise clients for Ernie bot has reached 85,000. Competition in China is heating up, Reuters writes. Recent data shows that rival domestic AI services, particularly the KimmiChatBot from a 12 month old Alibaba backed startup named Moonshot AI, are quickly catching up with Ernie. Bottom ErnieBot was visited a total of 14.9 million times across its app and website last month, while Kimmy had a total of 12.6 million visits in the same month. And Kimmy was growing much faster, with visits jumping 321.6% in March from February, while the number of visits to Erniebot grew more than 48%. Speaking of China, intel is following the Nvidia Playbook and launching a set of AI chips with reduced capacities that come in under US export restrictions for the Chinese market. The two chips are called the HL328 and the HL388 and are scheduled to come out in June and September. Nvidia also apparently has plans for three new China specific chips as well. Now, fittingly, we will end on this conversation about China and exports. As you will see, it has an integral role in our main episode which is about G42 and their new deal with Microsoft. However, that is going to do it for the AI Breakdown brief. Next up, the main AI breakdown. Hello friends. Quick note before we get into the main part of the episode. If you've been listening to the show for the last few months, you know we have been running an education beta. This is a new approach to AI learning that is hyper practical, focused on getting you actually using AI tools in minutes, not hours and certainly not days, based around video tutorials and companion challenges and projects that have step by step instructions that make it really easy to try out new AI platforms. And all of this is now culminating in the launch of a new platform which we're hoping is the most practical and useful way to learn AI that anyone has yet created. Go to be super AI that's besuper AI. Can't wait to see what you think. Attention AI breakdown listeners. Consensus 2024 marks the tenth gathering for all things crypto, blockchain and web3. However, importantly, this year's agenda will also dive deep into AI driven transformation, and the speaker lineup includes the leading minds and innovators at the forefront of this digital renaissance. Don't miss the Consensus AI Summit to cut through the hype to find where true transformation and opportunity lie. Listeners to this show can get 15% off registration with the code AIBREAKDOWN. Visit consensus2024.coindesk.com to learn more. Some of the folks will be at Consensys this year include Guillaume Verdon, AKA Beth Jesos, founder and CEO of Xtropic, as well as spiritual leader of the accelerationist movement Neil Stephenson, co founder of Lamina One, and Brendan Eich, the CEO of Brave Software. Again, go to consensus2024.coindesk.com to learn more and get 15% off registration with the code AIBREAKDOWN. Welcome back to the AI Breakdown. On the surface of it, Microsoft's new $1.5 billion investment in partnership into G42 is not all that surprising. First of all, G42 is a fast growing AI company in the United Arab Emirates. It's a region that's strategically positioning itself to be important in the AI space, and so it makes sense that Microsoft would want a foothold there. Second, in general, Microsoft is diversifying its set of partnerships in AI. It's trying to give its Azure business the greatest chance to integrate lots and lots of different AI solutions for lots and lots of different types of custom customers. However, when the broader context of G42 and the US tension with China when it comes to AI, the deal starts to look a lot more interesting. First of all, let's talk about what the deal includes. One aspect of it, as I mentioned, is the $1.5 billion investment to buy a minority stake in G42, and that's coming along with a board of directors seat. Microsoft's vice chair and president Brad Smith will be joining G42's board of directors. Secondary is clearly a cloud focus to the deal. Microsoft's announcement blog, for example, writes with the breadth of Microsoft cloud and its differentiated AI capabilities, the deal significantly advances G42 strategy of delivering generative AI and next generation infrastructure and services for a range of customers across financial services, healthcare, energy, government and education. Foundational to the partnership is G42's trust and commitment in Microsoft's cloud platform. G42 will expand its existing commitment to deploying Microsoft cloud offerings, demonstrating confidence in Microsoft as its preferred partner to enhance services and deliver value added solutions to its customers. With the partnership, G42's data platform and other essential technology infrastructure will migrate to Microsoft Azure. So again, all of this so far is very run of the mill. A big high leverage deal to try to win position for the Azure business line. One of the additional capacities that is being expanded in azure is that G42's Arabic large language model is now going to be available through azure. Language specific LLMs I think are going to be a big business. So again, all of this makes sense. One additional big aspect of the deal is that Microsoft and G42 will team up for a billion dollar fund for AI developers to boost the region's skillset. So those are all the basics and if you didn't know anything else that would be that. Why then are the headlines things like the Wall Street Journal's Microsoft Ups Ante in AI Race with China through Stake and Abu Dhabi Firm or the New York Times Microsoft Makes High Stakes Play in Tech Cold War with Emirati AI Deal for this we need to go back to the end of November when the New York Times ran an extensive story called Inside US Efforts to Untangle an Giant's Ties to China. American spy agencies have warned about The Emirati firm G42 and its work with large Chinese companies that US officials consider security threats. So the TLDR on all of this is that G42 has for some time tried to be neutral territory between the US And China when it came to AI. That was increasingly difficult, however, as they announced bigger and bigger deals. There was a $100 million deal with a Silicon Valley firm to build what they boasted would be the world's largest supercomputer, agreements with pharmaceutical giants like AstraZeneca and then in October, a partnership with OpenAI. However, behind the scenes, American security agencies were convinced that G42 was hiding just how much it was working with China, wrote the Times. US officials fear G42 could be a conduit by which advanced American technology is siphoned to Chinese companies or the government. The intelligence reports have also warned that G42's dealings with Chinese firms could be a pipeline to get the genetic data of millions of Americans and others into the hands of the Chinese government. The CIA even produced a classified profile of Peng Zhao, the Chief Executive of G42, who was educated in the United States and renounced his American citizenship for an Emirati one. So there are really layers of geopolitical issues here. One is that just in general, the Middle east is an area where the US has been trying to limit China's influence. Broadly speaking, the US has put pressure on Middle Eastern allies to deny Chinese efforts to build military bases. However, over the last couple years, the fight has taken on a distinctly technological edge, including especially but not exclusively artificial intelligence. Now, when the New York Times published this piece, it had been clear that discussions around these issues had been ongoing for some amount of time and that the discussions were not at the company level, but were between high level US officials and their Emirati counterparts. Now, there is a ton in here that's beyond the scope of this show. If you're interested, for example, in the genetic information piece of this, definitely go check it out. But suffice it to say, this article represented the bigger issues with G42 starting to come to a head. A couple weeks after that piece, G42 started to indicate that they had gotten the picture and that they were focusing Firmly on the U.S. we Count, for example, Read, a major Emirati AI company has picked a side in the U.S. china tech war. However, for some in Washington, that was a little bit too much lip service. In January, Wisconsin Republican Mike Gallagher, the chairman of the House Select Committee on the Chinese Communist Party, started pushing for a Federal investigation into G42's work. The House panel said that it had evidence that G42 works with blacklisted Chinese entities and that its CEO operated a network of Emirati and China based companies to support China's military and intelligence services. In his letter to Commerce Secretary Gina Raimondo, Gallagher wrote, multiple US companies that develop and sell export controlled technology and products maintain extensive commercial relationships with G42 and its subsidiaries to include Microsoft, Dell and OpenAI. Without restrictions on G42, the hardware and software developed by these US companies are at significant risk for diversion to G42's PRC based affiliates, many of which support China's surveillance, state and human rights abuses. With that continuing pressure, G42 started to move from statements to actions. In February, for example, the company announced that it had sold its stake in Chinese companies, including TikTok's owner, ByteDance. The $10 billion technology investment arm of G42, called the 42X Fund, told the Financial Times in February that it had, quote, divested from all its investments in China. While it didn't give specifics, two other sources pointed to the ByteDance shares as one of the major parts of the sell off and indicated that its stake in TikTok had previously been worth $100 million or more. So that is the background here. G42 is at the very center of the proxy war for the US's AI battle with China. And that brings us back to this Microsoft deal. The Wall Street Journal writes, relations between the US and the UAE have been strained in recent years over Abu Dhabi's growing ties to China and its technology firms. The agreement between one of America's biggest tech players and Abu Dhabi based G42 signals a pivot toward Washington for the Gulf State. Now, interestingly, not only was this deal approved by Washington, it seems like the US government actually had a stronger hand than that. The New York Times DealBook wrote a piece this morning called How Washington Played AI Matchmaker in which they argued that the White House quote laid the groundwork for the deal. That article shared another piece of the deal which didn't feature prominently in Microsoft's announcement post, which was that G42 had agreed to strip out components and equipment from Chinese companies like Huawei from its systems. Commerce Secretary Gina Raimondo apparently played a role as a lead negotiator in talks with G42. What's more, the New York Times writes, the investment is a collaboration between business and Washington. It arose out of dialogue between US officials and tech executives last year over how to encourage business transactions that deepen American interests in important regions and technologies. By joining the G42 board, Microsoft will be able to audit G42's use of its technology. So in many ways it feels like as G42 was trying to orient itself towards the US, it perhaps became a requirement from the Americans that they take on some sort of significant ownership stake from a US company. Given pre existing relationships, Microsoft was a likely candidate. And then a couple months later. Here we are with this deal. This may seem all kind of small to you, but we haven't really seen this type of collaboration between the Washington D.C. establishment and the private sector in this way around not just business objectives, but geopolitical objectives for basically as long as I can remember. Sure, Washington has often had an opinion about the moves that tech companies make, but this level of engagement actually playing a negotiating role, actually setting terms of a deal, shows just how central artificial intelligence is to us. Strategic priorities. Really, really interesting stuff, and I'm sure not the last time we're going to see this type of DC private sector collaboration. For now though, that is going to do it for the AI breakdown. Until next time. Peace. Sam.