before we head out of here. the futures are down across the board. as you can see, it's fed day. it's nvidia day. the brackets need to be in by i think tomorrow, don't they? >> oh, my gosh. >> got to be in by thursday. they've got the two playing games. i have nothing to bet on. make sure you join us tomorrow. "squawk on the street" coming up right now. we created a process for the generative ai era, and one of the most important parts of it is content token generation. this format is fp 4. that's a lot of computation. 5x, the token generation, 5x, the inference capability of hopper seems like enough.

but why stop there? the answer is it's not enough. >> jensen huang unveiling the new generation of nvidia's ai chips yesterday at the company's developers conference. welcome to "squawk on the street." i'm david faber. live from the new york stock exchange. jim cramer is in san jose, california. he is at nvidia's gtc. he's going to have a live and exclusive -- it's both -- live and exclusive interview with jensen huang. that will be in the next hour. of course, we or very much looking forward to it. a half hour from now we get started with trading. let's give you a quick look at futures. we are in the red. so perhaps a lower open when we begin trading. it starts with nvidia's potential ai game-changer. the company unveiling its latest, more powerful artificial

intelligence chips which ceo jensen huang says are twice as powerful for training ai models that the current gpus. the fed's march meeting kicking off today. the bank of japan raising interest rates. no longer are they negative in that country. and why elon musk thinks investors should want to keep taking ketamine, saying it's beneficial for tesla. we start with nvidia and the company's new ai chip lineup. there's jim. i throw it over to you. give me the highlights. you're about an hour or so from sitting down with jensen huang. jim, what struck you as most significant? what is most important today that was said yesterday in terms of how nvidia trades? >> okay. i think it's very simple. this new blackwell chip, it's smarter than 100 fabers. yep, that's what it looks like.

basically he's saying we're not smart enough. the machines are smarter. it's finally happened. we have chips that have learned so much that they have common sense. they can do things that i know you may be concerned about but i think are major for society, productivity and waste, build factories, sing. they can do anything. they can paint. they also make it so the world is a better place. that's what we're going to be talking about. >> all right. let's get more specific on the blackwell itself which again he made the point, it's the name of a platform. it's not a chip. first of all, if you can, jim, explain to people what that means. get into some detail if you can. i'm sure you will with jensen, obviously, during your interview. give us a sense as to the leap that's been made. >> sure. i think you hit on something that's really important which is it's not just a chip. it is a platform. what does that mean? it means it's filled with software. what's that mean? it's recurring revenue.

it's not one off. you get in the rico system, stay in the rico system. a lot of people say it's a defensive move. i say it's an offensive move. what they can do is use video, text, many different languages and understand. you can imagine initially it will understand between a zebra and a horse. in the end it will know more about zebras and horses than we do. i have to use an animal imagery because this is something -- jensen spent a lot of time saying how it understands video. if it understands video, it can understand things we desperately need in society. rebots to do a lot of jobs. we need them. to have people be able to design things never able to design before by, yes, david, i'll use it many times. a digital twin means you basically can build anything, do anything with an nvidia chip and actually have it made in real life. factories -- the largest market

left, he described it as the final frontier, and yet that's what blackwell will do better than anything. >> god, i don't know where to start with you, jim. you mentioned robots. i know he had a bunch on stage with him, right? >> yes. >> i've talked about them in the past. obviously, tesla's, for example, what they've been able to do and how advances are being made in terms of the manipulation of hands, for example. what did you see up there? what are the expectations? has that become a real product line in some way? >> i caught up with jensen after yesterday when i saw him later on at an analyst con fab. he told me there was a green robot and an orange robot. the green robot wanted to hog the show, came up, very aggressive. the orange robot was recalcitrant, stayed back, stubborn, did not do what jensen wanted. the problem, david, with this is that actually may be true. robots are thinking. >> the robots are having

personalities of their own and not doing what we want them to do. we're already there. already the will smith movie? it didn't take long. >> yeah. look, there was no explanation for why the orange robot hung back other than a decision to make it so it wasn't going to obey jensen. it was the first time they connected two robots. he said this to me. david, i know some of it is tongue and cheek. is there any doubt that you and i can't easily be replaced by the green and the orange robot once they have learned what we talk about? do you think we can do better than they can? >> i don't know. i think we will be replaced and/or all of our consciousness will be uploaded to the cloud anyway, and they'll download us back into some sort of robot. >> right. >> we'll be doing this show for a millennium. >> how much do you think the rebots cost to talk? >> i don't know. >> i mean once they're bought.

i would sayer zero. the cost of ownership of a faberless anchor, they call them fabulous, but it's really faberlous. >> maybe they have the little nuclear reactor in them. >> everybody actually hopes -- david a great point. the grid is being taxed. 5% a year is growing because of data centers. when you go into a data center, what do you see? you see super micro racks. thanks so much for offering 2 million shares to take advantage of being out of the s&p and up more than 250% for the year. you see some cadence product, synopsis product. but it's dominated by nvidia. they don't have the blackwell running. right now using the h100. jensen said over and over again this iteration is much bigger. one of the things i'm going to be talking about, he spent really quality time talking about apple. no one talks about apple. it's like fight club, the first rule about apple is you don't

talk about apple. he did. maybe that shows you, david, this man you're seeing actually has more power than anyone out here and can talk about a apple partnership. >> he fills an arena in a way like it's a religious experience. so, yeah, he's clearly moved -- he's the center -- >> he pointedly told me in the end, he said, i'm a computer engineer. it's really hard for me to be who people want. he is not a rock star. he doesn't want to be a rock star. he's a computer engineer that has great partners and is taking advantage of his product. i think the misunderstanding of the woodstock ai is just -- it's just so ill-advised. it's not what's happening here. it's not woodstock. >> when the movie is done, when we look back and see who was at the center of whatever is going to happen, it's going to be him. he's the central character. >> it's not going to be country joe and the fish. >> it's going to be him. it's not going to be timcook.

maybe it will be elon. it's probably going to be him at this point, especially when it all comes crash down when the robots won't listen to us and they basically have their hands on our necks. jensen is going to be the star of the move spri. >> will you stop with the skynet already? i talked you up yesterday talking to jensen. i said, i've got this guy who is my partner. he just thinks of skynet. >> i don't know him. i wish i did. i have incredible respect for everything he's accomplished. i want to come back to it. >> the power consumption, we talked about that a lot. >> they talk about this being more powerful, but consuming less power. i've seen confusing things, jim, originally in terms of flops. be serious, 2.5 times more flops

than current h series. overall has knifed times better flops and 30 times better performance. ultimately we'll consume more power. the computing power is way, way. multiple is more, but where are we on this? >> first of all, david, it's not cal flop. it's a tera flop. the interesting thing is pro lodgics is building these, they go are awe round the country and find where the cheapest power is. some are ready. the grid will be ready by 2030. i know that's a little late. but nuclear is the answer. we don't build nuclear. >> these blackwells will require -- ultimately will require a lot more power than we have available right now. >> yes, it will. i think, no, that's not true. there are places that can provide the power at a

reasonable price, but eventually yes. if we're growing at 5% a year, this grid that we have is made for growing about 1% a year. started growing at 5% a year which is too fast. i think it's a major issue. jensen speaks about waste, he speaks about climate change, speaks about what all this power could do to the earth in terms of making the ocean hotter. it's all bad. >> these are significant concerns. that said, the hope perhaps is that generative ai and slshg or when we get to true artificial general intelligence, they'll be able to figure it out. maybe we'll see some solution that we aren't capable of. >> you know who is figuring it out? it's not going to be humans. it's going to be machines that are trained. using inference, you put this question in in english. it comes back in english. it has the answer because any one of these blackwells is

smarter than any thousand people you have at your company. by the way, if you have 500 people at your company and you need 500 more and you can't find them, you plug in a blackwell. >> going to be able to do that as well. >> a fairly decent level. that's been a very important development. you're saying now it can actually design the chip -- going to design its replacement chip, the blackwell platform or design what's coming next? >> over and over -- i will use the words of jensen huang because he's a diplomat. we simply aren't smart enough to do what this machine, when it actually has the data does. that's the real issue. it can do faster and smarter than we can. you might say, wait a second, it can replace it. jensen says no, it's great productivity device. you've got to get right with jensen. you've got to get right -- >> i'm going to be listening to

every word he says in your interview. i hope you get to some of these. >> thank you very much. they're $10 million. they're $10 million. jensen is sitting on a gold mine. >> so are our viewers, by the way, if they own their own home. i don't know if you know that, jim. >> can i talk about ketamine for a second? >> no, you can't. you can't talk about it. >> it's a real drug. why can't he take it? johnson & johnson is ketamine. >> this guy doesn't even know when he's not in trouble. >> when we come back you can talk about ketamine because elon musk had that interview with don lemon, talked about his drug use of ketamine or at least referenced it. we're going to get to that and some of the things he said about the roadster. more "squawk on the street" coming right at you after this

.

at pgim, finding opportunity in fixed income today, helps secure tomorrow. our time-tested fixed income suite, backed by over 145 years of risk experience, helps investors meet their goals. pgim investments. shaping tomorrow today.

elon musk defending his taking ketamine. this is in an interview posted with don lemon yesterday. the ceo suggesting it's been beneficial for investors in his companies. >> tesla is worth about as much as the rest of the car industry combined, from nothing. so that's pretty good. we have the best-selling car on earth last year. from an investor's standpoint, if there's something i'm taking and it's working, i'm going to keep taking it. >> you pointed out it is a medication that's widely used. the "wall street journal" has come after him about this. not unexpected it would be part of this conversation. >> here is a man who admits he

has depression, admits he takes a drug that has helped millions of people. for that's he's pilloried? i know it's hiyperbole defendin what he did. there would be people never brave enough to admit they take this drug. but this drug is a lifesaver. >> did not watch the entire interview, but i know a good amount of it also. dealing yet again with many of the issues that musk and i discussed last may in terms of his need to share his every thought on so many different issues, many divisive issues and why he does it. jim, he did offer some things about the new roadster. if you have a reason why the stock was up yesterday, but maybe one reason, he did talk about it, sort of having

rocketry stuff in there, a combination of tesla and spacex technology to create something that's not even really a car. >> he said over and over again it's a tech delivery system. i think one of the things that's happened, if you speak to steve sc scherr, the used carvajal you of tesla has gone up in the last month. that's often a floor. that's allowed elon musk to raise prices. that's something that's counterintuitive to the narrative that we think. it might be true. the writing off the ev is a little premature. as you and i both know, we'll all be driving evs by 2030. i think this is the bottom in the stock. i've been waiting for the prices of the used cars to bottom and they are. i'm excited about the roadster. i'm excited about what he's doing. i think we need a refresher.

this is a big refresh. it does matter tremendously. i've got to ask you about, in your interview, did you detect the importance of his mood in terms of creating things? >> what do you mean? >> well, i mean, this is one of the most creative, if not the most creative person of our era other than jensen huang. why should we care that he needs a legal substance to be able to create things that you and i can't create? >> yeah, that's a valid question. it's not necessarily one i even think is really that worthwhile of great discussion to your point. sadly, there are a lot of people who deal with mental health issues in this country, and you want all of them to get the best treatment they possibly can. >> i worked with j&j on this to see what it can do for people and it's remarkable. i hope people recognize from this man that he saved a lot of

lives. i don't think he meant to, but he did. >> as nor his continued posting on x and some of the stuff he says there, that's a little more problematic. a little more problematic. >> should revert to "princess bride." >> there are definitely some connections strangely with henry ford sometimes. >> don't say that. that's a very dangerous concept. >> go back to 27 million circulation. >> i know. the independents. obviously the protocols. we all know his history. it's kind of interesting. >> interesting. interesting. >> that's one word to use here. >> that's horrible. that's horrible. >> i hope that's not true. >> get your "mad dash" ready. we'll get back to talking stocks. just in case people don't know,

jim is going to have an exclusive with jensen huang in the next hour right here on the "squawk on the street" television program. [kids screaming and playing] ♪ ("tosca, act ii: vissi d'arte" by maria callas) ♪ ♪ (orchestra del teatro alla scala, milano) ♪

♪♪ [water splashes] ♪♪ [kid laughing] ♪♪ [water splashes] ♪♪ ♪♪ [ripple sound] [laughing] [water splashes] [continue laughing] ♪ ("horn house") ♪ ♪♪ [chuckle] ♪♪

>> there are the laggards on the nasdaq so far. well, in extended trading obviously. it appears amd going to be amongst the big losers when we begin trading. marvel right behind it. opening bell just a few minutes away.

♪ ♪ this is not just another e-class. because it evolves with you. it adapts to you. engineering. it is the first e-class made just for you. for you. for you. this is not just design because your e-class... it adapts to you. it recognizes you. understands you. empowers you. energizes you. feels you. it evolves with you.

the new e-class. ♪ ♪

let's do it. another cross-country "mad dash." two minutes to the opening bell. unilever is saying to accelerate growth, it's the old shrink to grow, jim. they want to jettison ice cream. >> yeah. breyer's and also klondike, not just ben & jerry's. i want to know as nelson peltz did, when are we going to get rid of knorr.

they aren't going to have any food when this is over. they want to be in consumer products non-food which is smart given the fact that we know there are drugs from eli lilly that makes you not interested in ben & jerry's. >> you think we'd be better off -- >> as beauty and well-being, personal care, home care, nutrition, that's it. get out of food entirely. >> they're going what proctor did, raised proctor's margins, also raised its pe. it's a very smart move. unilever needs to do this. i question why they didn't do the whole food. why do they have to do one part of the food. nelson peltz in the board room pushed for all of food. david, nelson peltz, i'm sure you'll have something to say about him later. >> we'll talk about disney a little later. we are the largest per capita consumer of ice cream in the world, just so you know.

did you hear me, jim? >> phenomenal. [ cheers and applause ] >> wow, those people have game. they're gamers those people. >> very loud here. opening bell, of course, take a look at the realtime exchange. more red on the board. [ bell ringing ]. >> maybe that will drive the stock price up. >> gentex, supplier of windows and mirrors to the automotive and aircraft industries. jim, where do you want to start? >> we have to start with nvidia. >> i think almost every company we talk about, whether it be a technology company or not technology company wants to get ride with nvidia. the reason is because everything moves fast or smarter. it's a do-over of everything we have. see how i have this computer. you have to type things in.

who needs a keyboard when jensen is done? i think what we're seeing is everything that's electric has to go and change. everything is digitized. david, anything that can move, a robot moves better than we can. robots are human-like because everything has been designed to be for humans. so the robot should look like us. >> stop talking about the robots that look like us. it's freaking me out. >> no. it makes sense. we can't have robots look like something else because everything is designed for humans. robots have to look like humans and be humans. but smarter. you can teach a robot that coffee is hot and the robot can teach you that it's starbucks, not pete's. >> if nvidia is so great, jim, why is the stock down? >> because, david, it's been up.

>> i'll get a split here. i promise you i'll get you a split of nvidia stock. >> why are you even confusing people as though a split means something of any fundamental value. i'm just asking after the big presentation yesterday and what seems to be very positive developments for the company, obviously headed up by the blackwell platform. i'm just curious of your thoughts about the stock being down over 2% as we get started. >> i said don't buy the stock around the conference because the conference is not about raising earnings. it's not about forecast meaning. it's about ten-year thinking. can you imagine ten-year thinking when you're filled with a room of people with ten-year thinking? they're trying to build a dynasty, not trying to build a good first opening drive. it takes time. the stock anticipated this. the event itself is not what

we're -- it's a new way of thinking, a way of thinking that's better than what we do. it's not threatening, not meant to do anything other than being more productive and smarter which the world needs. it's going to produce less waste in the end. the cost of ownership is lower than what we have. it's faster and smarter. how do we not want drugs that can be developed in maybe three months instead of three years when we have just a short time of patent protection. therefore, the drug companies don't attempt to go after hard things. they won't go after the hardest illnesses because of this. they shy away. it takes too long. this is the end of that. this is the end of factories that waste so much that we have to question, question what they do for the planet. it's the end of a lot of the ways we think. it's all positive. it is. it's positive. >> okay.

>> say something. >> i'm listening to you, thinking about it. i'm not sure i completely share that. i'm trying to think of the world you're describing. >> do you think it's positive or negative? did you favor the seamstress over the loom? where were you on the steam engine? >> fulton is buried right next to hamilton, isn't it? we'll have a lot of free time. we'll have a lot of free time. i'm not sure what we're going to do with it all. >> we will be so much smarter and we'll be helped. we can feed it nvidia. it recognizes everything. >> we'll have robots waiting on us. >> the robots have common sense. they have common sense. >> when are they going to decide that we're not essential? when are they going to decide why are we around? >> we program them. we tell them whether we're

essential or not. they don't tell us we're not essential. >> you think they'll have intelligence greater than our own? >> david, there is no such thing as -- >> it's people, jim. >> could you do edward g. robinson. >> i can't put him from the ten commandment and then put him in soil -- >> it's hard to change everything. >> you and i will continue, until we're replaced by robots, we'll continue talking about the future. amd shares down 3%. what you saw yesterday, impacts on amd. >> look, amd has chips. you need the package. you have to have thought added to the chips. you can't just send a chip to some company and say, hey, go figure it out. you have to have jensen's team come, explain to you the software platform, why it has to

be integrated. lisa su is incredible. her chips, she will tell you, rival the speed of the h100. they can't compete with what we saw yesterday. it's the thought that comes with it. instead of talking about training and inference, it's thought. this produces thought. this is an ai factory. advanced growth doesn't have an ai factory, as much as i think lisa su is great and her stock has been unbelievable. think of a factory of thought. it's not a chip. it's a factory of thought. >> got it. okay. >> we want to talk disney as we get closer and closer to the big vote. the proxy fight, nelson peltz seeking a seat. the news this morning, got support from -- george lucas supporting bob iger. sold lucas to the company back

in 2012. it was half cash, half stock. you got some 37 million shares. not sure how many he continues to hold, but would make him one of the larger shareholders if he's anywhere near that level. larger, in fact, than pearl md der. what they're claiming as their overall $3.5 billion investment in the company. >> what's the stock done since nelson decided that bob iger is not doing a great job? >> depends where you measure it from. >> we had jeff sonnenfeld on. he talks about nelson being a net destroyer. sorry. i'm not going there. i just think that's wrong to say, because you have to talk about what he does in the board room and whether he helps in the board room. that's why i thought the jim stewart piece was so good. >> a very positive piece for

peltz. the jim stewart piece in the times. >> one of the most thoughtful journalists of our time. >> he's written some really good books. i'll tell you that much. >> lucas told cnbc creating magic is not for amateurs. when i sold luis cacfilm just over a decade ago, i was delighted to become a disney shareholder because of my long-time admiration for its iconic brand when he returned, i was relieved. no one knows disney better. >> wouldn't it be interesting if george lucas has money with t tryian? >> you think he does? >> well, stringer things have happened. >> yeah, they have. i have no idea. we're getting closer. all the big guns are coming out here so to speak. obviously i had jamie dimon last week provide me that -- he provided a quote in support of

bob iger as well. he's not a shareholder, per se. >> does disney do work with jpmorgan? >> yes. we pointed that out many times. you were away, doing relaxation. >> i didn't relax a single day. i was studying nvidia every day of my vacation. >> it was rare for dimon to come out in that fashion and support iger in the way he did. of course, we do have to point out there was a relationship between the two. every day is going to have something here. we'll keep watching it. yesterday apple and alphabet were the big outperformers, jim, on the story in bloomberg that said gemini might be used to power the generative ai offering from apple when we get it. backing off a bit today, any further reflections from you? >> i think a lot of people feel

that no matter what happens apple won't have it fast enough, good enough. you and i both know that apple doesn't release anything before its time, so to speak, to quote the late orson wells. i think vision pro got a real boost yesterday, yet people didn't talk about it. i'll talk with jensen early on. i found a new use for the vision pro that i didn't realize, which is to be able to sell cars, sell homes, make people realize you can put yourself in a car and put the -- i don't know, the blue ridge mountains and what a sunset looks like while you're driving a car. that's what jensen can do for apple. i think apple needs that. people are searching for uses of the vision pro. this is a great one. >> you are talking about driving a car. come on, man. think about everything you've talked about the future, and you're still talking about driving a car? >> a technology vehicle. listen. you ever go in a lamborghini at

185 miles an hour like my wife did? 185 miles an hour on a regular street. yes, i like cars. >> i haven't, actually. i did go in a porsche once, 192, down a freeway in dallas, texas. my friend kyle bass was driving that. yeah, that was a little crazy. >> you know we're on tv right now. >> we're on tv. >> you want to talk about the fed? you want to talk about the bank of japan? a fed meeting begins today. tomorrow the press conference. you always advise against the pros. that's been your take for a while. >> press conference is ridiculous because people raise hands. can you please tell me when you're going to cut rates. the next guy says, hey, can you give me a sense of when you're going to cut rates? forget it.

>> i want to get my number right for my model. it's the same thing. >> exactly. that's what analysts do, what reporters do. that's why they should be replaced by orange and green robots. >> we don't get any other colors? we only get to choose between orange and green? >> right now we just have orange and green. i'll ask jensen about more colors. i'll get everything straight. >> how long are you going to get with him? >> i'm not coming from the amd conference. i'm here. >> what is behind you? what is all that stuff? >> there are hundreds of companies that are partnering with nvidia. nvidia is a provider of thought, machines. everyone wants to think smarter. you think deloitte doesn't? there's cisco. i do see many companies that want to partner. that's what this is about. this is in many ways -- i'm

going to use the pedestrian term. it is a trade show. it's a trade show. but it's the most important trade show in history. got it. >> looking forward to the sit-down. it's getting close. are you getting nervous? >> are you nervous? i've prepared for this all my life. >> all your life one preparation -- >> this is what i prepare for. this is a 20-year process. i'm going to be talking about two types of stacks, the computer stack and, yes, the stack from denny's which has bananas, ice cream and two eggs and bacon. that's the stack we're talking about, because, yes, jensen worked at denny's. he washed dishes which is below being a busboy. i got demoted from busboy to dishes. i know this. at the block and cleaver by mr. fox. i hated him. >> okay. >> go get ready.

>> you've got to get ready for your interview. >> bviously, a lot more on "mad money" tonight. we'll see you in your special hour the next hour with your special guest, jensen huang. >> time for the bond report. you look at treasuries, 4.314 on the ten-year. two-year still hanging in there above 4.7%. we're right back. clear

new projects means new project managers.

you need to hire. i need indeed. indeed you do. when you sponsor a job, you immediately get your shortlist of quality candidates, whose resumes on indeed match your job criteria. visit indeed.com/hire and get started today.

coming up, we're going to have an interview with the ceo of cleveland cliffs. they were the cover bid on u.s. steel deal and have been very much active in trying to prevent nippon steel from acquiring the company. the ceo joining us right after this. your skin is ever-changing, take care of it

with gold bond's age renew formulations of 7 moisturizers and 3 vitamins. for all your skins, gold bond. they're waiting for you. hey, do you have a second? they're all expecting more. more efficiency. more benefits. more growth. when you realize you can give your people everything, and more. thank you very much. [applause] ask, "now what?" here's what. you go with prudential to protect, empower and grow. with everything you need to deliver, you guessed it... more. one more thing... who's your rock? learn more at prudential.com

my name is oluseyi who's your rock? and some of my favorite moments throughout my life are watching sports with my dad. now, i work at comcast as part of the team that created our ai highlights technology, which uses ai to detect the major plays in a sports game. giving millions of fans, like my dad and me,

new ways of catching up on their favorite sport. grayscale,. we're keeping an eye on shares of u.s. steel, up a bit today, down nearly 17% since president biden weighed in on nippon steel's to buy the company for $5 million in cash.

competitor cleveland cliffs for its part said it would consider a bid for u.s. steel's assets or some of them if the deal with nippon steel were to fall by the wayside. joining us a cnbc exclusive, cleveland cliffs ceo lourenco goncalves. lourenco, good to have you. you've been a one-man wrecking crew in terms of kind of bringing the heat on this deal, especially given your relationship with the union versus nippon steel's. why has it been so important to you to try to prevent this deal from actually happening? >> yeah, actually, good morning, david. a pleasure to be with you this morning. actually, my position all along with you constructive. my offer was done in a way that was prepared to make things to go through smoothly and that's why i started with the union, the usb. the usw has contractual rights

and i respect that, and one thing that u.s. steel did not respect. that's when the problem started. at this point, it's just a matter of waiting. it's not if, it is when. >> what to you mean it's not if, it's when? you don't believe there's any possibility that nippon is going to be able to work a deal with the unions in its favor to be able to close its transaction? >> i have full conviction this deal between nip been steel and the union will never happen. we have an agreement with the usw. we have the right to bid on their behalf. we understand the successorship clause. that's something clear that nippon steels into clue about. we are going to prevail. the president of the states has already spoken. it's no longer whether or not the government will approve. the government has expressed

their decision in terms of who should keep ownership of the assets and it's not a company from japan. japan is not a friend. japan -- in -- >> japan would see that pretty differently, lourenco, in terms of saying they're not a friend. it seems hard for some people to imagine it could be found to be a national security risk from a cfius review. how do you think this deal is going to fall apart then? obviously, the president has indicated he wants the u.s. steel to be a domestically owned, but it's got to be through the cfius process, doesn't it? >> how are they found to be a national security threat? >> cfius will do their work and cfius will generate a report that will land on the desk of the president of the united states and then the president can accept, reject, or modify the report. basically the president can do whatever the president wants to do. and director brainard of the

national economic council expressed opinion and the japanese did not understand. then the president of the united states himself spoke loud and clear and apparently nippon steel, that started with an embarrassment with m&a, wants to transform this thing in a diplomatic problem between the united states and japan. so if that's the case, let's do it. let's play. >> well, what influence have you had on the biden administration? obviously, dave mccall who runs the united steel workers union, we had him on yesterday. he clearly seems to have a strong influence on the president. have you been behind that as well? >> we have a very, very strong partnership with the usw. it's unusual. it's not common. it's a new thing that i believe it's about time for investors to take notice.

we can work with organized labor in a constructive way. we don't need to be fighting. we need to be working together. that's what middle class is about. that's what union jobs, good paying middle-class jobs are about and the president of the united states is fully committed with this type of agenda. that's what we have. >> you know, but there's -- there are those who believe this is simply your attempt to get nippon steel to agree you sell you the blast furnaces, take big river, give you the assets you want, you waged this to get them to the negotiating table and then you've pulled the lever in terms of politically to get the biden administration to come against the deal. is that a fair perception? >> it's not. it was the perception until the president of the united states, president biden, spoke loud and clear that the ownership of the assets should stay with the united states. i would be able to do that until

they, meaning the japanese, pro voting u.s. government to react the way the president reabcted. different from the japanese i pay attention to when the president of the united states speaks. there's no more negotiating a deal between cleveland cliffs and nippon steel. nippon steel is out. even the shareholders vote will not do any good. >> they don't think they're out, lourenco. you know that. they think they have a deal in place. they have a contract. probably get a positive shareholder vote in april. you seem to be saying something very different than what they would say, we should still be able to negotiate a deal with the unions and why wouldn't the unions want to in some way versus being left at u.s. steel with no other company owning it? >> the problem with that, it has hit the iceberg. at this point there's no point in pretending that the ship is already -- is still sailing. the ship is sinking.

it's sinking slowly and we are very patient. we're not going anywhere. we're going to wait until the deal is unraveled and pick up the pieces on the other side. >> what does that mean, picking up the pieces on the other side. >> u.s. steel has no future because of the way they were disrespected, the way they treated the union as a third-class citizens, there's no situation that nippon steel doesn't like but nippon steel is not going to buy and u.s. steel continues as an ongoing concern. someone is going to be at the other side. you bet i will be there and be able to -- >> you were there during the process as well. you bid $54 a share in cash and stock. obviously, not far from what was the winning bid. one of the concerns, lourenco, the antitrust concern, the fact that a combined company, cliffs and u.s. steel work have 95% of iron ore production, 100% of

blast furnace production among antitrust issues. how do you get people comfortable with the antitrust risks that made u.s. steel uncomfortable? >> yeah. u.s. steel brought this in order to alienate the union. that's what happened. at the end of the day, competitive, it's competitive with furnaces. united states steel built big river to compete against us. nucore is building a plant in west vas to produce. steel dynamics built a plant in central texas. we have to compete against imports and we have to compete against other materials like aluminum. we will handle that when the time is right with the doj and we have a plan and these things are easily resolved when you know what you are doing. >> would you be willing to give shareholders hell or high water

protection against the high water risk? >> i'm not going to negotiate with you, dave, but i know what takes to get a deal done. >> again, to come back to it, they have a contract to get bought for $55 a share in cash. even though the president has said what he has said, nothing is changing that. i mean, i come back to this question of what gives y the confidence that this deal is a fail? >> at the very least the clock is unforgiving. coming june of 2025, there's no deal, the deal expires. another thing that can happen instead of having president biden, we have president trump, he gave me personal assurance he will block the deal at the very beginning if not on day one for sure, at the very beginning of his tenure. >> president trump -- former president trump told you he would block any deal as well? >> yes. >> he said that to you? >> he said that. >> have you spoken to president biden? has she given you the same

assurance? >> i have not. my partner dave mccall is in direct contact with the white house and we have assurance from the biden administration as well. >> lourenco, to be continued, but certainly appreciate your taking time with us and helping explain your side of the story. thank you. >> thank you. appreciate you having me. >> lourenco goncalves, ceo of cleveland cliffs. that saga continues. good tuesday morning. welcome to another hour of "squawk on the street." i'm sara eisen with david faber, live from post nine of the new york stock exchange. carl has the morning off. want to show you a live shot right now of nvidia's developers event in san jose, california. the company launching its next generation ai chip. several wall street firms raising their price target on the launch. just moments from now our own jim cramer will be sitting down exclusively with nvidia's ceo jensen huang. it's an interview you cannot miss. take a look at stocks, though,

here on day one of a fed meeting just the dow higher of the major averages and that's thanks to some strength in nontech names. that's the story today. the s&p 500 it's down a quarter of a percent because communication services, technology, those are your two worst performing groups. you have notable strength today in utilities, energy, staples, financials catching a bid, consumer discretionary even higher as well. but with the tech trade not working, that weighs on the s&p and nasdaq down 0.7%. nvidia the biggest point drag on the qqq. apple and microsoft remain higher. take a look at treasuries. six straight days of higher yields leading into the fed decision. today they cool off a little bit. the 10-year note yield 4.3, the 2-year 4.7% compared to where we were a few months ago. david, the -- i made three charts for the fed week and the fed meeting.

here's one from bank of america and merrill lynch, about the projections where rates will go and every fed member gives then. the last time september. the median prediction they would cut three times this year. he shows a chart how it's been a ceiling for the pricing of fed funds futures where the market thinks the fed is going to go. so there's been all sorts of adjustments based on hotter inflation numbers of how many cuts will come into the market. it sits where the fed last predicted. the surprise would be if the fed goes from 3 to 2, for instance, fewer cuts, in today's dot plot. >> you pointed to that as a possibility. >> it's a tough call. another surprise would be more hawkish rhetoric out of fed chair powell -- >> at this point i wonder if it would be a surprise. i guess yesterday i think it was cathie jones from schwab expects more hawkish tone. any number of people seem to

think that's a real possibility. >> how hawkish right of the tone. so far his line and most of his deputies' line has been the disinflation trend is still in place, we need to see more progress, data that continues to affirm that. tore shin of apollo put out a chart while we made a target on inflation we're stagnating around 3%. this is could be problematic for the fed. economists will look at that chart and say that's not problematic for the fed. look how much it's come down and it will continue to come down. that's the nuance that fed chair powell needs to interpret and we'll see how concerned they are or how willing thoey are to tolerate these levels of inflation. >> put a chart up at the press conference and ask a question, how do you view that chart? >> i wish. i -- i know, at one point i

wanted to join congress to ask good questions -- >> you would be so good. it wouldn't have been worth it. >> the other factor to watch, we haven't talked about this a lot, does enter the story is the balance sheet because the fed told us they will begin discussions on the balance sheet and how to taper the qt, tightening they've done, on liquidity. here's the balance sheet chart. it got to -- >> that's tapering? i mean look at that. >> no. no tapering yet. it peaked around $9 trillion. >> right. >> after they basically doubled it during covid. >> sorry. but that's -- that's -- >> that's quantitative tightening. >> that's the tightening what you're making fun of. >> yes. >> barely everything. >> it's -- it's significant for them that they have been able to do that. the problem is, and it's very technical, kind of plumbing issues, they're watching this reverse repo facility which is a good measure of liquidity in the financial system and banks have been drain reeg serves from that and they want to avoid any kind

of market fallout or market turbulence vupts of that. there was turbulence in 2019. >> to go to the chart. what was our high? >> $9 trillion. >> now we're? >> below $8 trillion. >> okay. that's it. >> well, it's not it. they're going to continue to do more, but expect comments about how and when they might start to taper the tightening. >> taper the tightening. >> expectations now from economists are in the may meeting and start to see the reduction of how much they let run off the balance sheet. >> why? why taper the tightening? you saw in 2020 we were at half that. >> because it's used as a tool that goes in conjunction with tightening policy to fight inflation on higher rates and that's starting to taper as well as they move into rate cuts as they have fought inflation and so it risks draining liquidity from the market at a time it could cause a further disruption

without getting too technical into the reverse repo facility that's their concern, and we have seen signs that facility is being drained and that's why some of the more vocal fed presidents, ones who have experience on market desk, are starting to talk about we need to slow it down in terms of the tightening. i'm going to be watching for those headlines. >> we'll come back to some of the other ones as well. for now we've got to go to nvidia's ai developers conference and get to jim cramer. he joins us with a very special guest. jim? >> thank you so much, david and sara. i am with jensen huang, and we are at what some people are calling, jensen, the woodstock of ai. but isn't it much more than that? isn't it about a change in everything we do when it comes to digital, when it comes to creating, comes to thinking? you're changing that. isn't that what we're doing out here? >> yeah. this is an incredible conference. this is nvidia's developer

conference. everything that we do starts with software. everything we do starts with software. everything we do is in service with all the software developers solving these difficult algorithms. we are represented by $100 trillion of industry here. health care is here, financial services are here, manufacturing, industrial, automotive, climate, tech, you know, holy cow, communications is here, consumers are here. >> but people always think of you as hardware. you're talking about a different platform, a system that, frankly, may be unassailable from competitors. once these companies get involved with you they're going to stick with nvidia. >> it's specialized way of doing accelerating computing. what we do is this, this is the observation a long time ago, 30 years ago we observed that the cp is good at many things but some things it's surprisingly ungood at. parallel things, things that you

could distribute across a large number of processors, so what we did is we added nvidia to a cpu. we connected it to a cpu, off load the work that cpu is not good at and we run that work insanely fast. surprisingly, that work that the cpu is not good at represents 95% of the time that is spent in computing. we off load that 95% of the time and we run it 100 times faster. >> you're talking about a total do over of all technology! >> yeah. >> you're talking about everything -- our country is building -- new -- new plants using old technology. that's the case. it's wrong. >> well we should build amazing semiconductor plants here, and we would be more than happy to build all kinds of chips here. it's very clear that in the future, that general purpose computing, it's like a general purpose almost anything, you know, general purpose instrument

of any kind it's not very efficient. there are many types of things we want to do very efficiently, computation of mathematics we want to do efficiently. as a result of doing it efficiently you drive the costs down and use less energy. one of our computers -- our latest generation -- this is the chip that goes into it. this is the largest chip the world has ever seen. this is beyond the limits of physics. we had to invent some new technology to make it possible -- >> how many -- >> it's 208 billion tris transistors. >> in that. >> in this tiny part in the middle sdmoom how much does that cost? >> this will cost y, you know, , $40,000. >> how much did you spend? >> the r&d budget is like $10 billion. >> not million. >> billion dollars. >> you deserve the right to be able to recoup that and you're doing it. >> we're going to do our best

job. this computer -- >> the name of that. >> called the blackwell computer. >> named after a very -- >> yes. a mathematician. >> yeah. >> really terrific mathematician. and this computer here will replace thousands of general purpose computers. this is the part that's incredible. in fact, what's amazing is that the cables of connecting last generation general purpose computers, the cables of connecting them, cost more than the price of one of these computers. the amount of energy that we save is incredible. mega watts. because of this, we made it possible for the computer to write software by itself. it is so insanely fast the software can write -- the computer can write its own software and we call that artificial intelligence. >> if that's the case why do we still need us? >> well, we still have to guide the software. we have to create the algorithms

such that the computer can go write software. and that algorithm is called deep learning. yeah. it's really quite a remarkable thing that happened in the last 12 years. >> if we ask it questions, inference. >> yeah. >> it speaks our language? >> well, if you ask it a question, first of all, it not only recognizes the words, but it understand your meaning. it understand the meaning because -- >> nuance? >> sure. you can give it a -- say -- first of all let you read this book, read "moby dick" and then ask you questions about it. first it goes off and reads it and takes a flash -- >> does it understand whyishmail is driven? >> absolutely. it read the end of the story. the read the end of the story. not only that, it's read a whole bunch of other stories and so it understands the context of the conversations, but it also has encoded within it a lot of

things that it's already read from society. >> okay. but you're describing something that's different from earnings per share. you're describing wonderment. you're describing it creating something that can replace trillions of dollars of what we don't need anymore, do it faster, do it more productively, do it cleaner. everything has to be replaced! >> there's a lot of waste in the world. there's a lot of waste in the world. oftentimes we can't chase it down. of course there's a lot of wasted energy used in doing computing and now with accelerated computing we can make it a lot more efficient. there's a lot of waste in just about every single industry. the challenge is that we've never been able to use a computer to understand the information of that industry. one of the things that's really exciting is, we've been able to sequence genes, but never been able to understand what it means. we've been able to -- >> so we didn't understand the

proteins do? >> we can begin to understand what a protein does. >> if we do that, then we can do drug trials in 60 days instead of six years. >> that's right. >> so companies will taingle the illnesses they can't afford to tackle. >> that's right. >> at the minimum you have to do trials on people and things like that, but we can reduce the time that it takes to go through the entire search space of drugs and proteins and targets and that search space is just gigantic. it's impossible for humans to to do it. we can now, because computers with artificial intelligence can understand the language biology, we can sort through that a lot more quickly. >> how about the last frontier? can it understand a factory? >> the last frontier, we have to teach it to understand physical things. it has to understand that when you drop something it falls to the ground but doesn't go through the ground. you have to understand that mechanical hinges work in a

particular way, and so this mechanical hinges work in a particular way or the laws of physics. this is no different than word sequences and sequences of sentences turn into paragraphs and so on and so forth. the computer can understand, can learn to understand physics and mechanical things -- >> can it also -- >> once it does that we can understand how a factory works. >> can it understand [ inaudible ] diet coke? >> it's common sense. >> it's common sense. >> it's common sense and, therefore, of course. of course if you order fries, you should also, you know, recommend some diet coke. >> does it bothszer you in the end we're trying to figure out whether it's a trillion dollars for nvidia, $300 billion for amd. are these questions pedestrian versus the ten years you've worked to get here? >> we do very different things. nvidia is an accelerated computing company. if you look at the things we do. we build the chips, the systems, the networking, and so on, the

entire data center practically. all the software that goes into it. and then we sell it in parts. the reason we sell it -- that's what confuses people. they think nvidia is a chip company because we sell everything in parts. >> right. >> the reason we do that is so that our customers can integrate nvidia's technology into their data centers however they like. everybody's data centers are different, everybody's systems are different, and so when we build up the whole thing, we make it work, but we sell it in parts so that fits into the, you know, nooks and crannies if you will. >> if you can make it up like that and have the software, why would there be any other semiconductor companies? >> there's lots of -- jim, the world of semiconductors is gigantic. we serve one nish called accelerating computing and artificial intelligence. this is an important nish because it's the foundation of computing as we know it going forward. nvidia is a data center scale company, full stacked software company, and we design the entire computing system.

we sell it in parts so that everybody can enjoy nvidia -- >> one last question -- why is nvidia a $2 trillion company? >> oh, gosh. that's a tough question. first youof all, there are seve things i appreciate about the work we do. the single most important instrument of humanity is computing. now we have computers that could understand information of all different kinds. the impact so -- to the industries are enormous. $100 trillion worth of industry are here. the impact of the work that we do to all these -- afford all these industries is incredible. >> you can take your 2% of the $100 trillion. >> thank you very much. >> now we're going -- we have to toss it back to david and sara and then we will speak for "mad money" tonight. thank you. >> jim, thank you. and thanks to jensen as well. fascinating to listen and, of course, as jim mentioned, the

interview is going to continue and if you want to see it you have to watch "mad money" tonight at 6:00 p.m. so there will be a lot more of that conversation. that ongoing conversation. sara, i found it fascinating. because it's not -- it wasn't a business conversation. it's a society conversation. i mean he didn't even -- "moby dick" he didn't finish his thought but was saying if you read "moby dick" in a millisecond. >> and jim followed up understand their characters and motivations because it would have read that book and other books. i think the examples are what stuck out, fries with coke, common sense. a peek into our future and also i thought a plain and simple explanation of what nvidia does and whether wall street either fully grasps it or does not about how he calls it an accelerating computing -- >> accelerating computing company, different than amd.

>> you're going to hear that more. that starting to get more use. accelerated computing. >> more than building chips. they build the data center and platform. >> plblackwell is a platform, n just a clip. he did holding that up, again 208 billion transistors. >> when you think about texas instruments and where we started 50, 60 years ago, 208 billion transisters, it boggles the mind. they had the conversation a bit about the language of biology and what that's going to mean in terms of the service space and how quickly it can find things out and quicken the discovery process for new drugs. >> now all the companies are rushing to put out press statements and also do partnerships to partner with nvidia on these game-changing

technologies. he honed in on health care and pharmaceuticals, but there's a lot of potential with a lot of different industries. >> jim and i last hour were talking about the robots and everything else. it does come with some risks. he doesn't talk about it as often. >> jim asked him why do you need us. >> to program the algorithms. >> until they can do it. >> stacy bernstein analyst joins us for instant reaction along with kristina partsinevelos there at the developers conference. stacy, the stock is down a little bit today. do you think it's just -- i mean it did add a trillion dollars in market value this year. what do you think the market is reacting to? >> the hype into the event was massive. i've never seen anything quite like this before. they sold out the s&p center. it was more reminiscent of a taylor swift show than an event.

i think we were expecting to hear news flow on -- and expect everything else for the new gpu architecture. we heard everything we were looking for. i don't know that we heard anything that we weren't necessarily expecting. looks like a great chip. everybody expected that. that's why it's selling off a little bit like not under performing the broader semiconductor industry by very much. it's taking a breather. >> kristina partsinevelos, what are your takeaways from hearing jensen and being there? >> for just from the interview the biggest thing he just added a price tag to the blackwell chip. i'm waiting to confirm with the pr team for nvidia on which particular chip because like you talked about, blackwell is an ai platform on there there's going to be various chips provided over the next year or so, so jensen held up a blackwell platform, and he said between 30 to $40,000. which to me i was expecting it to be a little bit more and stacy, too, i've seen his note, a lot of analysts have weighed

in saying between 40 and $50,000. at least 30% higher than the previous h-100 clip. if that's the case, it could be good for margins if demand keeps up. that's a strong takeaway. yesterday yeah there was -- it was a sold-out event, celebrities, all that, but what we did -- the message we got from nvidia is that they're not just the hardware company, they're providing the full stack solution. jensen mentioned that. that will make switching costs harder should you choose to leave nvidia in a few years and go to the in-house options that a lot of mega cap tech players are trying to work on or the likes of amd or intel eventually, who knows. >> right. the creation of the ecosystem. stacy, do you see that as well, and how important does that become in terms of the models that you build to figure out what this company will be earning in years to come? >> the ecosystem is hugely important. they do everything, they do the chips, they do the hardware, the

software, they do the networking, they do everything. there isn't a single other company on the planet that does everything like to the extent, everything that's around the clip. it creates a tremendous amount. people have been talking about the software systems cuda, they've been doing this well over ten years. every other competitor recognizes that, every other competitor has a crash course now to try to improve their software offering. the problem is that you're fighting against a player that has been entrenched over a decade. there's a massive install basis that has millions of developers that can program on nvidia's platforms. it's a very, very broad moat hard to crack. most of the other players are discovering that. >> stacy, just something i would love you to be able to explain to me and our viewers, power consumption. we talked about it in the broader context of data centers but the introduction of this platform was accompanied by saying that it consumes less power for more computing.

but explain really what's going on here, because i've got a little bit lost. >> yeah. so on a per workload basis, every architecture they introduce is much more efficient. i think they said something that could be as much as 25 times the efficiency of operating. take those with a grain of salt. like fine, they're going to throw out like good, big numbers, but it should be more efficient on the per workload basis. on an absolute basis, they use a lot of power because you're cramming more computing to the data centers and dragging a lot more power on an absolute basis. it's like something that may become an issue like bradley for the space how do you get the absolute amount of power required into these data centers to do the compute. given the amount of compute they're doing on a per workload basis much more cost effective. we need to get the power in there. >> understood. now i get the relationships. >> how does this change the competitive landscape? he started to talk about how different of a company nvidia

was from an amd. they're both lower today by the way but have done very well. how does this change and potentially leapfrog nvidia's capabilities here, the new blackwell clip? >> well, this shows that nvidia, we talked about the full stack solution almost to the point where nvidia is trying to become a cloud service provider competing with its own consumers. they mentioned aws, microsoft, google, all of these guys that are providing or going to be using the blackwell chip but at the same time the same companies are creating their own in-house chip. i call the relationship almost frenemies. it's interesting how nvidia is trying to pivot continuously away from just providing solely hardware and the entire ecosystem and that's going to provide a strength. but to your point of how they can switch or crack that nvidia code i think cost is going to be a major factor and that's where some of the other players may come in saying they're going to offer cheaper chips but switching costs may not be the easiest. and then stacy and you talked

about it, power. it is definitely going to be a problem and that 25 x that nvidia provided is not necessarily accurate when comparing specs but it is something notable and something amd has to consider with their new mi-300 chip why you saw a sell the stock reaction for amd given the specs we received from nvidia. overall i think it's power, cost, cost is a big one. everybody wants an alternative to nvidia and there is none affordable at the moment. >> stacy, to that point and we see amd shares down 6.5% supporting your point about the fact ta it's not trading worse than the group nvidia. what's good for nvidia is not good for all competitors? does the ecosystem benefit. >> if the opportunity is as big as some are suggesting it might be, that's good for everybody in the space. at the same time as nvidia moves, their performance is going up and up at a much more

rapid pace then most of their competitors at this point have not presented road maps that can keep up with that. i want to push back a little bit. your comments on cost. the chip cost is certainly important, but that's not really what matters at the end of the day. you think about performance thor dollar total cost of ownership. the cost of the chip is only one aspect of this. most of the competitors are offering cheaper chips but the performance differential is such that it is probably still cheaper like on a per workload basis to use the nvidia parts. you can look where the dollars are flowing, right. amd suggested they might do $3.5 billion worth of ai scale this year. numbers for nvidia are bumping up on $100 billion. you can sort of follow where the money is going. it's clearly going mostly in one direction. these are the reasons why. >> yeah. it's been a demand story -- >> wouldn't you say -- >> go ahead. >> no, i'm just wondering,

inferencing, do you feel like that could change the game? you're spending a lot of money on the training done for these mega cap players, but inferencing is the next step. nvidia saying they're getting 40% of data center demand. does that cost not play a role when it comes to inferencing where others players could fill that gap? >> it absolutely does plays a role in everything. people view inferencing as this like low-cost or like low performance. inferencing is not -- it's a spectrum. right. you have something running in your hand on your phone, but like a lot of these like large language foundation models the compute requirements for the inference are massive, absolutely like cost efficiency and power, absolutely. >> my final question is just about demand. it's about supply with all this demand. there's a lot of it and what we've learned about whether they can keep up and how long it will take. >> yeah.

nvidia has suggested their -- on the earnings call a while ago their current generation is getting better. most of the constraints around capacity, we have more supply coming on line. they suggested their new products would continue to be supply constrained, and it does look like the demand is there. at least for blackwell, i think they announced -- he had a chart, 50 different customers on it. there's demand versus hopper two or three customers when hopper launched. there's clearly demand and i think for the new stuff, the supply constraints are likely to be there. the silicon area, these are bigger chips. blackwell is two big stuck together that has more memory and more packaging capacity. i think it will be tight for that even as the current generation supply eases up. >> thank you both for joining us to react to that interview and talk about nvidia which is down a little bit today after a

stunning rise this year and last. >> yeah. >> having an impact on the overall market i would say. >> yeah. as perhaps amd is as well. far less in terms of overall market value. still at $2.1 trillion. that's the market value of n nvidia. bring in mike santoli. >> it fit into the themes in the markets, though. first of all this small decline in nvidia after the massive build up is almost like the market hit a bull's eye on what to expect and had been priced in. even from the morning after nvidia's last earnings report, less than a month ago, the stock had added almost $200 billion in market cap. after the news mostly fulfilled expectations, what's next. it fit into a broader chase of the momentum to the upside. part of the story with nvidia is everybody racing to capture the fundamental opportunity. you see how much earnings

estimates have gone up. that's one piece of it. then you have sort of the speculative momentum overlay on that where it becomes just this daily game of, you know, kind of computer ping-pong and we trade $100 billion in worth of the stock many days for no particular incremental reason but because it's the game and then the adjacent stocks get caught up in it. today you have just a little bit of a reset on all that stuff. the momentum trade has been probably over extended starting a week or two ago. it has come in a pretty benign way. the market has not buckled under it. you have things like crypto, bitcoin, 10,000 below its recent high, super micro doing a secondary offering getting hit. i think there's a little bit of a, you know, froth being skimmed away but also just people are using the opportunity to kind of have a little mental reset of how much has been already built in to these prices. so far the rest of the market is handling things okay. we'll see if that can continue

to be this somewhat seamless rotation, this commodity move is interesting. the rest of the market is kind of in wait mode on the fed. >> i wonder how much has been priced into the rest of the 499 stocks as it related to what nvidia is talking about and the future of ai, how it's going to change our economy, change company's productivity, their earnings power, whether that's in the market right now? >> i can't imagine it's in the market in any precise way. it's on people's minds. people recognize that we could be in for a bit of a productivity boost just across the board. that would fit with the mid-90s scenario? terms of a soft landing kind of with an additional enhancement of this productivity boost. i just think it's too hard. >> how can you figure that out? >> the investment phase. we're kind of hustling to try to get the capacity where it needs to be so the companies can figure out what it can do for

them. >> can read "moby dick" in a millisecond. >> add the economic value attached to that. what economic footprint are we kind of displacing here with some of these capabilities some. >> professors. they're incredibly important. >> everything turns into customer service is more efficient, targeting ads is more efficient, you know -- >> they did talk about the factory floor. >> exactly. >> what you could do there. >> and manufacturing. >> gets more efficient. >> i mean analysts -- >> but it's all hovering out there in the cloud and even -- he was looking at the forward estimates for nvidia itself, okay. the earnings and revenue system for fiscal 2026, next fiscal-year, january fiscal year, is, you know, we're at 28 times that number. it's like $30 a share. the fiscal year after that, $33 a share, 10% grower. it could be another double. but right now the market is only willing to say after next fiscal

year, we get a 10% bump and, you know, we're not going to necessarily try to price that in. 26 times fiscal 2027. it's not crazy. it's a little above. it's not crazy. >> let's go back to easier just parsing the level of hawkishness from jay powell. >> right. >> trying to figure out -- >> plot some dots. >> counting the dots. >> that's where we are. >> blackwell will tell you whether at some point let you know. >> predict the fed moves. >> yeah. >> your whole portfolio for you as well. that already happens. >> exactly. people are already have those models and running them as we speak. >> all the time. >> perhaps we don't talk enough -- >> and champions -- >> de -- >> 50% of the time. >> so many. >> citadel and renaissance. enormous impact. >> right. >> another report we'll talk about. let's get to the story. >> but after the break here george lucas is backing bob

iger, the ceo of disney, in the proxy battle between disney and the activist investor nelson peltz. at'll give you more details on th. don't go anywhere. hm? you! your business bank account with quickbooks money, now earns 5% apy. 5% apy? that's new! yup, that's how you business differently. [♪♪] your skin is ever-changing, take care of it with gold bond's age renew formulations of 7 moisturizers and 3 vitamins. for all your skins, gold bond.

business. it's not a nine-to-five a proposition.. it's all day and into the night. it's all the things that keep this world turning. the go-tos that keep us going. the places we cheer. and check in. they all choose the advanced network solutions and round the clock partnership from comcast business. see why comcast business powers more small businesses

than anyone else. get started for $49.99 a month plus ask how to get up to an $800 prepaid card. don't wait- call today. ♪

welcome back. i'm silvana henao with your cnbc update. president biden says he will immediately sign the new funding bill that's expected to hit his desk ahead of friday's government shutdown deadline. it comes after senate and house leaders said this morning they reached an agreement on the final set of full-year appropriations bills. brazil's federal police reportedly indicted former president bolsonaro today for suspected fraud on his vaccination records after an investigation found the covid-19 skeptic falsified his card in 2021. according to reuters, bolsonaro was indicted on charges of entering false data into the health system and forming a criminal organization. and the new york state comp troller's office says the average bonus fell by about 2% lastyear that brings it from $180,000 in 2022 to about $176,000 in 2023.

the comptroller attributed the declineto more people joining the industry. sara? >> thank you. film maker and hollywood legend george lucas is back bob iger in the proxy battle with nelson peltz. in a statement, lucas said, quote, creating magic is not for amateurs. when i sold lucas film i was delighted to become a disney shareholder because of my long-time admiration for its iconic brand and bob iger's leadership. when bob returned to the company during a difficult time, i was relieved. no one knows disney better and i remain a shareholder because i have full faith and confidence in the power of disney and bob's track record of driving long long-term value. we follow the back and forth that disney is bringing out the big guns with george lucas who is a big shareholder. >> he is a large shareholder. i'm not sure exactly how many shares he received roughly 37

million shares when he sold the company, half cash half stock deal. if he's held anywhere near that it would make him one of the largest shareholders if that is still roughly the number. >> at one point he was the second biggest outside of iger. >> not iger. iger has noefr -- >> so second biggest then individual. >> i would say on the trian slate of people who have come come out and endorsed peltz on the other side, it's a lot of the ceos who have worked with nelson peltz. i wept through a list this morning, mondelez, david taylor who led p&g and janis henderson -- >> have they endorsed him in this battle? >> on the site trian takes you to in terms of their fight they are quoted on there saying they have had very pleasant experiences and nelson knows his

stuff and has worked very well with them to enhance slair value on the board. it's a lot of that and a lot of friendly comments. schumacher, the ceo of unilever on there. companies that have worked with nelson. all the big ones you can think of, wendy's on there too. so that -- i would just say for an investor, for more institutional investor, i wonder if those comments resonate? because part of the fight that disney is putting up here is that nelson doesn't have any ideas and not going to work well. >> i can hear nelson with george lucas wouldn't you be happy with better performance. >> they're saying that. >> so we have to wait. >> yeah. we have a couple weeks to go and we have glass lewis yesterday. we talked about the proxy advisory firm. iss the biggest and most influential. we'll get that later in the week. >> the meeting april 3rd. >> yeah. the vote april 3rd. >> it's dramatic. >> both sides are confident, but

we will see. still ahead, results from our cnbc fed survey. how much is off landing is priced into this recorsein ocmaet?-ttg don't go anywhere. investment opportunities are everywhere you turn. but at t. rowe price, we're letting curiosity light the way. asking smart questions about opportunities like advances in healthcare. and how these innovations will create a healthier world tomorrow. better questions. better outcomes.

you know doug, ever since switching to workday you've been a real rock star. rock star? what do you know about rock stars? billy idol? i mean where's the skin-tight leather? my shoes are leather. where's the unnecessary zippers? that thing! billy, rock star is just how doug feels when he uses workday. thanks, rory. i'll show you rock star! be a finance and hr rock star. workday. for a changing world. billy idol just stole your golf cart!

welcome back to "squawk on the street." the federal reserve kicking off its two-day meeting on interest rates and fed chair powell's news conference. let's bring in steve liesman with results of our fed survey. >> sara, bond yields over the next couple years seen remaining around this 4% level, while stock returns, according to respondents of the fed survey, could be lackluster. yesterday's close at 5149, so the forecast, the average forecast, is just 5246 or about 2% higher. 5448 or about 6% higher for 2025. then take a look at the outlook for the 10-year, 4.33 current down just a little bit, maintaining a sense of that 4% level. now be careful here because this group has generally been pessimistic about stocks and not

necessarily right. take their forecast in that light. they are optimistic, though, on the economy with 52% believing there's going to be a soft landing as the most likely outcome for the economy. most think stocks are over priced for the soft landing scenario. see there 58% say over priced relative to the soft landing scenario compared to 24% in september which was a good call then when it wasn't as much over priced and now 27% saying under priced at 17% in december now 12%. so these are interesting numbers because back in september, their sense was pretty good about the market not being over priced. now they believe it is. senior investment strategist from philadelphia trust company says the fed is in no hurry to cut rates, current interest rateses are more normal than they have been. here's the rate outlook from the

respondents, 538 goes down for the futures market priced and then 360 or 356, the end of 2025, neutral rate at 310. there seems to be agreement between the markets and forecasters. a little bit better growth, a bit less fed easing. the question whether that is priced in and if bond yields remain high how much of a challenge that is to equities. it's not a tina world anymore. there are alternatives and bonds, if markets are fully priced there are alternatives. >> steve, your estimation of the consensus expectations for tomorrow is what on the projections and on the -- >> on the projections -- >> language from powell -- >> it's a close call. i don't know that anybody knows the actual bent of each of the 19 members to the decision

required to say it's going to end up being two rate cuts instead of three. it takes only two, but they have to be in a certain place to move higher as you know to get the median down from two to three cuts. i don't think it matters that much. i'm interested in whether or not they maintain the 2025 outlook which is kind of similar to our outlook on the survey here. a little bit of tweaking on the front end in 2024, but it's 2025 where the real action is and there would be relief to the economy. i think that's an important thing to watch. >> okay. we'll do it. thank you, steve. steve liesman. as we head out check out shares of unilever announcing plans to separate its ice cream unit which does include ben and jerry's and magnum that will impact around 7500 jobs. the stock up almost 3%. unilever has had headaches with ben and jerry's now and their semi independent board. new ceo wants to make moves. we'll be right back. stay with us.

power e*trade's easy to-use tools make complex trading less complicated. custom scans help you find new trading opportunities, while an earnings tool helps you plan your trades and stay on top of the market. e*trade from morgan stanley. meet ron. ron eats, sleeps and breathes hoops. and there's not a no look pass, double double, or buzzer beater he won't wax poetic on. ad nauseam. but oh how he can nail a software solution like the best high screen pick and roll you've ever seen. you need ron. ron needs a retirement plan. work with principal so we can help you help ron with a retirement and benefits plan that's right for him. let our expertise round out yours. ah, these bills are crazy. she has no idea she's sitting on a goldmine. well she doesn't know that if she owns a life insurance policy of $100,000 or more she can sell all or part of it to coventry for

cash. even a term policy. even a term policy? even a term policy! find out if you're sitting on a goldmine. call coventry direct today at the number on your screen, or visit coventrydirect.com. the future is not just going to happen. you have to make it. and if you want a successful business, all it takes is an idea, and now becomes the future. a future where you grew a dream into a reality. it's waiting for you. mere minutes away. the future is nothing but power and it's all yours. the all new godaddy airo. get your business online in minutes with the power of ai.

volatility for bitcoin. it's down about 6%. trading around $63,000. in danger of losing its monthly gains. all of them, in fact. checking in on the overall market, the dow is doing well but the s&p and nasdaq are being held back by weakness in technology stocks. notably, nvidia. other tech names, amd, tesla, intel down as well. we come back, nasdaq only down cts e owotr d most of the he seorargrn.

you founded your kayak company because you love the ocean- not spreadsheets. you need to hire. i need indeed. indeed you do.

indeed instant match instantly delivers quality candidates matching your job description. visit indeed.com/hire

by itself. it is so insanely fast. the computer can write its own software and we call that artificial intelligence. >> and that's the case, why do we still need us? >> we still have to guide the software. we have to create the algorithms such that the computer can go write software. and that algorithm is called deep learning. yeah, it's really quite a

remarkable thing that happened in the last 12 years. >> if we ask it questions, inference, it speaks our language? >> if you ask it a question, it not only recognizes the wordses but it understands the meaning. >> nuance? >> oh, sure. >> that was nvidia ceo jensen hueng with jim cramer. we can still do the algorithm. >> he was very casual about the answer. >> we still have some role here. and then the robots will just fully take over and that will be that. >> he did not say that. that's a faber ad lib. >> it is. that's a faber ad lib. >> robots were an important part of yesterday's presentation. at least they were on the stage as well. >> you heard the buzz word inference, which you're hearing a lot more.

jim used it, inference. the future, basically, of a.i. and workloads that they're seeing. we're hearing increasingly about it on things like conference calls for nvidia more than demand is. >> and you asked a key question earlier. we had a brief question with mike about productivity. i think that will figure prominently into what we talk about every day. when we start to see the use cases in enterprise for companies, where will it be used? how will it be used? and how much productivity will be in increased as a result? >> you and i talk to ceos about this all the time. the line has been, well, it's not going to amount to fewer jobs. it's just going to be different jobs. we have to retrain people. we're going to empower people. they'll have super powers now to do their jobs, which is a politically correct thing to say, but also it does make you wonder about certain back office functions. call centers, i feel, are a poster child. >> they are in trouble. >> for what we might not need anymore in terms of jobs. >> that's already under way. >> we haven't seen it show up in

the macro data yet. obviously, it's very early. that's interesting, too. it's why the fed and economists need to take a look at this from a growth perspective, from a productivity perspective, from a jobs perspective. there's a social fabric. the imf is looking at this and how they're going to compensate people -- >> who are sitting home doing nothing, more or less? >> it's happening. these people need to get working on it. >> we can't help but think about the societal impact, some of the things he's talking about. that does it for this hour of "squawk on the street." our live market coverage keeps going. . >> on "money movers." you know what's brilliant? boring. think about it. boring is the unsung catalyst for bold. what straps bold to a rocket and hurtles it into space?

boring does. boring makes vacations happen, early retirements possible, and startups start up. because it's smart, dependable, and steady. all words you want from your bank. for nearly 160 years, pnc bank has been brilliantly boring so you can be happily fulfilled... which is pretty un-boring if you think about it. you know doug, ever since switching to workday you've been a real rock star. rock star? what do you know about rock stars? billy idol?

i mean where's the skin-tight leather? my shoes are leather. where's the unnecessary zippers? that thing! billy, rock star is just how doug feels when he uses workday. thanks, rory. i'll show you rock star! be a finance and hr rock star. workday. for a changing world. billy idol just stole your golf cart!

of. good tuesday morning. welcome to "money movers." i'm sara eisen with mike santoli. former fed official vince reinhart, why he says the if ed will start cutting in june and continue for the next year and a half. everything you need to know from cnbc's exclusive interview with nvidia ceo jensen huang. >> a look at the housing sector and what new data means about being a buyer in this market