Tag Archives: G42

April 23, 2026 · 07:30

Bleeding on the spot

That is at times the setting, we tend to ignore it, we laugh, we giggle, and sometimes we cry. If it is your own body, you will likely panic. So as I saw Tom’s Hardware (at https://www.tomshardware.com/tech-industry/artificial-intelligence/cerebras-files-for-ipo-company-remains-unprofitable-despite-20x-revenue-growth) give us ‘Cerebras files for IPO — company remains unprofitable despite 20x revenue growth’ I tend to frown. There are settings with little profit (like the Big Mac for $1.95) which at 20 times still becomes a decent amount (all $6 of them), we get that other factors that remove profit margins, but when the setting becomes “Bleeding money at a rapid rate” it becomes a worry. You see, the business plan makes sense or is a hail Mary (not unlike the Macintosh Performa) this is an intentional setting I am giving, because that Hail Mary became the PowerMac and then the G4 and G5. These were the systems that put Apple on several maps and from there the big wins became visible. A Hail Mary that worked. But here we are given “Cerebras, the supplier of wafer-scale AI processors, has filed for an IPO for the second time after it cancelled such plans due to its ties with G42, an Abu Dhabi-based AI company backed by sovereign wealth fund Mubadala, last year. Financial results disclosed as part of the filing reveal that Cerebras appears to be one of the fastest-growing AI hardware companies right now. However, 86% of its revenue comes from two customers, and the company is bleeding money.” From this limited information I would gather that the business plan is highly likely flawed. And we are given that the 86% comes from just two customers (G42 and Mohamed bin Zayed University of Artificial Intelligence, MBZUAI). Now I would go with the Business plan, but there might be reasons for this and the settings that AI processors give could still be a solution if these two clients put in the considerable work (no critique on the two trendsetters). As we see that “The remaining 14% of revenue is generated by a fragmented base of smaller enterprise, government, and cloud customers, but none contribute enough individually to reduce Cerebras’ heavy reliance on its top two clients. More recently, Cerebras inked agreements to supply its AI hardware to Amazon Web Services and OpenAI, which will diversify revenue streams for the company.” But the larger option is gaining traction. Now for the most we can ignore the fact that they are American (which is at present never a good selling point), but they are also in Toronto and Bangalore. The issue is that they are no threat to Nvidia and they don’t need to be, the idea is that they could skim the market and take up traction pretty much anywhere. I reckon that they have done that, but there is the option that they could optionally feed data centers in China, Saudi Arabia and the UAE, if that works and they could get the first one in these places, they are likely to gain several other corporations and locations for implementation. The reasoning I have is that there are several sounds from customers that they have a lack of processors, so are they tapped? It seems so as we see “Cerebras has a massive $24.6 billion backlog (including the $20 billion OpenAI deal), which provides strong demand visibility. The company expects to recognize approximately 15% of this revenue within the first 24 months through December 31, 2027, 43% during months 25 to 48, and the remainder thereafter. Still, Cerebras warns that converting this backlog into revenue depends on the manufacturing capacity of its partners, infrastructure deployment, and power availability.” It makes me wonder why the quote “Bleeding money at a rapid rate” was given. So as we see “Cerebras recorded a $363 million gain from a change in the fair value (and extinguishment) of a forward contract liability: the company had a financial obligation whose value was reduced, which allows it to book that reduction as income. If the value was not reduced, the company would be unprofitable. In fact, Cerebras’ operating losses totaled $145.9 million in 2025.” But even so, as I see it (with my lack of economy studies) thematic doesn’t seem to add up and my mind goes back to the business plan. It is my simplistic mind that goes with the setting that Cerebras either has a product that works or they have not. If they do, the client has to pay and there are no freebees in this market, you do that if the product is shoddy, and the salesperson either deals with the buyer correctly, or they don’t. It is my rather simplistic setting of customer service, “we have a product and we would love to have you as customer, yet, our product is not free”, it will rock your world (for a price) and within that setting (and the right business plan) Cerebras should do just fine. As such I don’t get the setting we see. So as we are also given “Cerebras postponed its IPO plans in 2024 after a national security review examined its ties with Abu Dhabi-based G42 amid concerns about potential foreign access to advanced AI processors. G42 is both a customer and investor of Cerebras, which controls a 1% stake in the company that it acquired for $40 million in 2021.” This is an issue as it involves 50% of their customer base and what is this “potential foreign access to advanced AI processors”? Is this another American setting (not unlike their stance towards Huawei)? You see China is sized at 1.413 billion, as such it is over 4 times the size of the USA, the United States can either play nice or go down with the ship they are sinking themselves. Cerebras could go towards the EU as well as India and partially fund the data centers there and get longer lasting revenue, but that is almost the only options that are there. This market is getting saturated and it is not a market that has time and options for prima donna’s, this is my simplistic view. So as the article ends with “Cerebras has not specified an official fundraising target in its IPO filing, but current market expectations point to a roughly $3 billion raise. This is significantly higher than earlier $1 billion plans, which reflect the company’s rapid revenue growth and the scale of its AI infrastructure ambitions.” It also signals that the ‘bleeding effect’ is a temporary setting, depending on how the IPO evolves. Yet as I see it, the IPO has a lot less chance of being successful as long as the “Bleeding money at a rapid rate” vision is in place. But as I see it, enlarging their customer base precedes the need for an IPO, because no I matter how good the IPO is, it is facing slaughter when the customer base is set to two. But as I stated, my lack of economy might be the ruling red herring here.

And whilst I leave you with this article and a few hidden hints, I will go and look what happens to Cerebras before June, May it have a nice time.

Have an interesting day today (‘great’ is oversold too much, even by me).

The revolving question

That is at times in almost everything the setting. We might all go nuts about ‘mismanaging’ settings and I am to a certain degree not impervious to that setting. But after writing ‘The losing bet’ (at https://lawlordtobe.com/2024/12/08/the-losing-bet/) I started to mull things over. You see, people like Sheikh Tahnoon bin Zayed Al Nahyan are not stupid. But there is a dangerous calm as people are given the questions and are given ‘a kind of answer’ and Microsoft is massively adapt in setting the stage to THEIR advantage and I suddenly realised a simpler setting. When was the question asked of Microsoft ‘What is AI?’ And ‘What is the premise of what you call AI?’ With ‘What is the data setting of AI?’ In this I reckon that some eyes will open. We see all settings of Ai mentioned, but the clear definition and a comparison to the setting that Alan Turing gave us 1950, moreover together with John McCarthy gave us the Turing test. So how far did people dig into this part of the equation? You might disagree with me on my stance of AI and that is okay. We do not all see eye to eye on a whole range of matters. But in this, in a Texas Hold’em style of business poker it becomes increasingly important to set the stage of definitions and hold them up to the light. In that game Microsoft doesn’t get to spin out of the stage ad blame it all on miscommunication. In that stage Microsoft has to hide into the margins or come out into the light. The second stage is likely and very pleasing to my ego.

You see, when people are part of a $1.5 billion investment there are people who are not pleased with that fact and they will nitpick any document handed to them. One of the oldest settings was ‘What are the definitions?’ Was in older days the way to see what players were up to and that stage got a little lost in populism and ‘fast’ presentations appeasing to the spending player. You might think that it is Microsoft paying, but you would be wrong. The UAE and G42 are investing time and resources to make it all work and I foresee that players like Microsoft (not just them) are trying to play fast and loose with definitions so that they can bank the first agreements and then turn back and hide behind ‘miscommunications’ after that fact. Which is why we have the clear setting of definitions. As such making all players answer that question gives a first setting. You see, there is no AI at present and that comes out at that very start. And no matter how clever LLM’s and Deeper Machine Learning is, the setting becomes data and who is responsible of that data. Now we get different players out and in the full-grown light. People like Sheikh Tahnoon bin Zayed Al Nahyan will then immediately see who is endangering the security of the UAE and they have no sense of humour at that point. No matter how some see the ‘opportunity’ of a life time, the moment the national pride comes into view of danger, the UAE will demand clarity on matters and I reckon some will ‘trivialise’ matters and when you ‘invest’ $1.5 billion there is an issue with trivialisation (which is why I referred to a Texas Hold’em style). Now some will say that I am bluffing and I want to be ‘inserted’ as a possible player. You would be wrong. I do not want to be linked to a player like Microsoft in any way. Google, Amazon, Adobe, IBM and Oracle definitely, Microsoft not at all. As such I am not anti-American (a claim that was thrown at me several times in the past). I am anti-stupid (mostly) and when you start trivialising $1.5 billion I see you as stupid, and no matter what I think of Microsoft, they are not overly stupid. In some things yes, in other things (like playing black letter law stages) not that much.

But all that becomes moot when some players release the definition lists to all we will see how silly my thoughts are, because these definitions go through the entire project and there is no way they get changed unless all parties openly agree. Oh and before you think that this is a ploy. You might be right. You see, I do not know where China is at present ad I would live to find out. So what is better then Microsoft setting the entire definition list to paper and release it all? I reckon we will see a Chinese response less then 48 hours alter.

The revolving question is an almost needed stage because definitions on paper is what matters, if it isn’t written down it doesn’t exist. That has been a matter long before the Prince by Niccolò Machiavelli. I reckon it goes back to the days of Gaius Julius Caesar Augustus (63BC-14). So this setting was known for 2000 years and with all the turbo presentations and innuendo I get the feeling it got lost in the woodwork of it all. As such I thought it was a great idea to remind people of that.

Silly me, have a great day.

The losing bet

That happens, we make bets. We all do in one way or another. Some merely hurt our pride and/or our ego. Some deals hurt others and there are other settings, too many to mention. But Reuters alerted me three hours ago on a deal that will have a lot of repercussions. The article ‘US clears export of advanced AI chips to UAE under Microsoft deal, Axios says’ (at https://www.reuters.com/technology/artificial-intelligence/advanced-ai-chips-cleared-export-uae-under-microsoft-deal-axios-reports-2024-12-07/) is one that has a few more repercussions than you imagined it had. The global loser (Microsoft) has set up a setting where we see “The U.S. government has approved the export of advanced artificial intelligence chips to a Microsoft-operated facility in the United Arab Emirates as part of the company’s highly-scrutinised partnership with Emirati AI firm G42, Axios reported on Saturday, citing two people familiar with the deal.” Microsoft is as desperate as I think they are with this deal. They probably pushed the anti-China agenda and made mention of the $1.5 billion dollar investment deal. And as we are given “The deal, however, was scrutinised after U.S. lawmakers raised concerns G42 could transfer powerful U.S. AI technology to China. They asked for a U.S. assessment of G42’s ties to the Chinese Communist Party, military and government before the Microsoft deal advances.” And we are also given “The approved export license requires Microsoft to prevent access to its facility in the UAE by personnel who are from nations under U.S. arms embargoes or who are on the U.S. Bureau of Industry and Security’s Entity List, the Axios report said.” In this I have a few issues.

In the first there is no AI, not yet anyway as such the investment is going the way like water under a bridge. Microsoft knows this as such they are betting big and they have the US government backing them. In the worst case it will be the US government putting up the $1.5 billion themselves and with the anti-China sentiment that is a likely result from this.

In the second the setting that Microsoft is banking on is a loop setting with multiple exists. Yesterday the Financial Times informed us ‘OpenAI seeks to unlock investment by ditching ‘AGI’ clause with Microsoft’ (at https://www.ft.com/content/2c14b89c-f363-4c2a-9dfc-13023b6bce65) the events are piling up and as I see it Microsoft is on the edge if desperation. You see, it all hangs on the simplest setting that there is no AI (not yet at least). What we have is a setting with LLM’s and Deeper Machine Learning and it is clever and it is a ‘optional’ wholesome solution to a lot of paths. But it is no Artificial Intelligence. You see, as all the laws are part of ethics and ‘AI’ people look around and think that there is ‘awareness’ of solutions. There are not. It is all data managed, a somewhat clever solution to people seeking an aware-like solution in data and some kind of knowledge discovery mode. It all could be clever, but it is still no AI and at some point certain people will dig it out and I reckon the UAE will be ahead of it all. Microsoft and its Ferengi approach of ‘When you get their money you never give it back’ comes with nice loopholes. You think that Microsoft made the ‘investment’ now here is the cracker. There is nothing stopping Microsoft of putting it in a ‘bad bank’ approach and make it all tax deductible and then some. And when the “artificial general intelligence” (AGI) clause is dropped there will be all kinds of attention from all over the place and no one is looking at the details of whatever they consider AI and what Alan Turing clearly considered to be AI. When the people that matter start looking and digging the days of Microsoft will be numbered. Another bubble game created and now that they have ‘enticed’ the wrong kind of people they will want their pound of dollars. And as we are given “The Biden administration in October required the makers of the largest AI systems to share details about them with the U.S. government. G42 earlier this year said it was actively working with U.S. partners and the UAE’s government to comply with AI development and deployment standards, amid concerns about its ties to China.” And in that setting Microsoft decided to be the governmental bitch to say the least. And all these media moguls are so loosely playing along and what will happen when someone digs into this. They will play dumb and say “We didn’t comprehend the technology” and it wasn’t hard. I saw it months ago, if not nearly almost two years ago. And the media was stupid? No, the media goes the way of the digital dollar, the way of the emotional flame. So as the field opens, we see all kinds of turmoil with Microsoft claiming to be the ‘saviour’ all nice and kind (of a sort), but when you look at the setting, it is my personal speculated feeling that Microsoft wouldn’t have made this move unless they had very little moves left. And in this setting the one player is forgotten. China, how far along are their ‘designs’? And in all this what are their plans? We seem to be given the setting that it is all American, but as the media cannot be trusted what is the ACTUAL setting? I have no clue, but in a world this interactive, China cannot be far away.

And if there are people who disagree, that is fair, but the actual setting is largely unknown. So when we get to the last paragraph which gives us “Abu Dhabi sovereign wealth fund Mubadala Investment Company, the UAE’s ruling family and U.S. private equity firm Silver Lake hold stakes in G42. The company’s chairman, Sheikh Tahnoon bin Zayed Al Nahyan, is the UAE’s national security advisor and the brother of the UAE’s president.” Consider this small fact. Microsoft seems to be ‘investing’ all whilst the anti-China rhetoric is given. Do you think that anyone who is the National Security Advisor (of the UAE) hasn’t seen through a lot of this? So what was the plan from Microsoft? I am at a loss, but with the AI setting the way it actually is none of this makes sense. Do they really believe that Microsoft is any kind of solution in this setting? Simply look at the accusation that Microsoft has also been criticised for the perceived declining quality and reliability of its software. That is your partner in so-called AI? Just a thought to consider.

Well, you all have a lovely Sunday. My Monday is a mere 80 minutes away.

1 Comment

Filed under Finance, IT, Media, Science

Tagged as Abu Dhabi, AGI, AI, Alan Turing, America, Artificial General Intelligence, Artificial Intelligence, Axios, China, Chinese Communist Party, deeper machine learning, DML, Financial Times, G42, Large Language Models, LLM, Microsoft, Mubadala Investment Company, News, President Biden, Sheikh Tahnoon bin Zayed Al Nahyan, Silver Lake, technology, UAE, USA

May 14, 2024 · 21:00

Bully tactics?

I know, I added a question mark, because the Reuters article leaves me (at https://www.reuters.com/technology/uae-releases-new-ai-model-compete-with-big-tech-2024-05-13/) with questions. It is nice to see that the UAE is going all in for the AI generation and the Abu Dhabi’s Technology Innovation Institute (TII) is releasing a Falcon 2 series: Falcon 2 11B, a text-based model, and Falcon 2 11B VLM based system. However the US gives them scrutiny with the statement “American or Chinese technology.” As such the US is drawing battle lines in the middle east. Considering that we see “Emirati AI firm G42 pulled out Chinese hardware and divested stakes in Chinese companies before securing a $1.5 billion investment from Microsoft that was coordinated with Washington”. The danger for the US is that so far Microsoft hasn’t been delivering anything other that mediocrity. As such should this fall over the US will lose a hell of a lot more. Considering that the UAE is now BRICS, there is a chance that Microsoft (and a decent one) that Microsoft will go into its Delay, Blame and Miscommunication protocols. On the other hand I know NOTHING about either the Chinese or the American chip in question. The better setting is that UAE’s Falcon and Meta’s Llama, have made their code publicly available for anyone to use. As such that would imply the better creation of traction with the population. We are also given “Al Bannai said he was optimistic about Falcon 2’s performance and that they were working on “Falcon 3 generation”” In the end this was the first I heard about the Falcon, so we will be getting a lot more soon enough and I should look into the TII and what else they have cooking. You see “a vision-to-language model that can generate a text description of an uploaded image” has added bonus regarding yesterday’s blog. A simple scan of the boarding pass, or luggage tag could imply that booking a new tourist could be done in seconds. I have never been able to check into a hotel in under 5-10 minutes. Take that amount and multiply it by 53 million. That is an annual saving of 505 years. And that is only one country. That and a few other blogs I have written about shows you the essential need to upgrade tourist systems. The question. Then becomes whether such a system will have an Arabic setting and the UAE will have a system that benefits the UAE, Saudi Arabia, Egypt and Indonesia. All places with growing tourism at present. That is a setting not to be ignored. And the west? Well that remains the question, they will see the benefit of upgrading son enough and when Hilton, Marriott, Hyatt and others see that time benefit, they will all come on board regardless of where that chip originated.

Time will tell who wins that caper in the end, because the $1.5 billion investment doesn’t last long when you are confronted with Delay, Blame and Miscommunication tactics (if that happens).