News Intel's next CEO could be Lip-Bu Tan, a former member of the board who reportedly clashed with Gelsinger about strategy

[Admin]

Intel begins to shortlist potential candidates to the CEO position.

Intel's next CEO could be Lip-Bu Tan, a former member of the board who reportedly clashed with Gelsinger about strategy : Read more

 

[YSCCC]

ah right, getting a investment focused person over engineer personale is one of the dumbest move move for a tech company where it took years to let tech mature

 

[pclaughton]

A company the size of Intel has an economy almost like a country. The next CEO's first 2-3 years will be entirely the result of Gelsinger's tenure, so it seems crazy to cut the guy now.

 

[JarredWaltonGPU]

A company the size of Intel has an economy almost like a country. The next CEO's first 2-3 years will be entirely the result of Gelsinger's tenure, so it seems crazy to cut the guy now.

And by extension, nearly everything that happened during Gelsinger's stay was basically the result of stuff that happened before he even took the reigns. It's the board of directors doing the usual thing: It wants profits, now, and when profits are way down because of bad decisions made a decade ago, the current CEO gets to take the blame.

I think there are worse options than Lip-Bu Tan, but I'm not sure who would actually be a lot better. He has technical chops and he also has a lot of connections. The latter could very well be more important than the former as a CEO.

 

[JamesJones44]

Intel's board is largely made up of members with software backgrounds and investment bankers. That's not faith inspiring when looking for an outside CEO to turn around a semiconductor company.

 

[AndrewJacksonZA]

smh.

...Lip-Bu Tan... has a lot of connections. The latter could very well be more important than the former as a CEO.

He just might need them. Not just for company deals, beating back ARM, and governmental contracts, but I also just read in the news of mineral export bans from/to various countries because politics will politic, and unfortunately for tech companies, politics is extremely relevant to their bottom lines.

And more importantly, shareholders' returns on their investments.

 

[JarredWaltonGPU]

Intel's board is largely made up of members with software backgrounds and investment bankers. That's not faith inspiring when looking for an outside CEO to turn around a semiconductor company.

I will defer to my cohort Paul on this one, as I don't really know much about various CEOs / executives. Here's a convo regarding Lip-Bu:
-------------
Lip-Bu Tan could be the savior, dude knows how to run a fab. He is royalty in the chip world. The only person more revered is Morris Chang. He understands the way the fab networks work.

He has been involved in all kinds of startups and chip companies spanning decades. He gets it. He is an industry guy. So the rumor is that Lip-Bu Tan left the board to run a committee studying how, when, and if Intel should spin off the fabs. He is a big proponent of that, at least according to said rumors, and had disagreements with Gelsinger about that strategy.

So, if Lip-Bu is in fact their pick, one could assume they would plan to spin off fabs. As CEO of Cadence, he interacted with the entire third-party foundry industry. Like, he knows the people, he has connections, he knows how to do things. He would be a superb pick.

https://www.imec-int.com/en/press/jim-morgan-and-lip-bu-tan-receive-imecs-lifetime-innovation-award
-------------

And my personal take? Even someone with all the right connections and background may not be able to succeed, given Gelsinger failed. And spinning off the fabs is still messy business, could impact CHIPS Act funding, etc. Even after AMD split off GloFo, it was years before the albatross around AMD's neck due to prior contracts went away, and the same would likely happen with Intel.

 

[UnforcedERROR]

but I also just read in the news of mineral export bans from/to various countries because politics will politic, and unfortunately for tech companies, politics is extremely relevant to their bottom lines.

Eh, this isn't a political argument so I'm sure you're fine. It's very relevant to the space. China's bans can potentially impact this sector tremendously, and it's driven by political decisions (which they've openly admitted). I think it's a relevant point to make, personally.

 

[93QSD5]

A company the size of Intel has an economy almost like a country. The next CEO's first 2-3 years will be entirely the result of Gelsinger's tenure, so it seems crazy to cut the guy now.

They aren't stupid, maybe we are the ones who are dumb.

Think about it. Maybe they fired Gelsinger after he was on a path to fix things. They basically steal all the credit and keep all the future profits near the top, by putting a greed-head into the seat, right as the recovery might start to look likely.

Sounds like planned robbery to me. The only losers are the customers

 

[subspruce]

ah right, getting a investment focused person over engineer personale is one of the dumbest move move for a tech company where it took years to let tech mature

and we have seen it before cough cough Brian Krzanich

 

[phead128]

A company the size of Intel has an economy almost like a country. The next CEO's first 2-3 years will be entirely the result of Gelsinger's tenure, so it seems crazy to cut the guy now.

The Board gave him 4 years (Pat served 3 years 10 months) to pursue his aggressive "5 nodes in 4 years" plan. We saw 20A got cancelled, 18A is only 10% yield. The original plan is not 5 nodes in 6-7 years, at an unsustainable burn-rate of $17 billion per quarter! Let's Pat be accountable to his own timeline.

 

[vanadiel007]

I honestly think they will have a hard time recovering from this downswing, no matter who they are.
A lot of consumers who normally trust and purchase Intel will think more carefully, considering the CPU debacle and the unknowns about Intel's future.

Even their GPU sales to consumers will be a though sell, because who is going to purchase that not knowing if driver support will still be available in 6 months?

These are some serious concerns they have to address, and they do nothing besides lay people off and spin of business units. They need a future roadmap, and fast...

 

[-Fran-]

If they hire him, let's see if there's anything in the rearview mirror for him.

Regards.

 

[Kamen Rider Blade]

Intel's board is largely made up of members with software backgrounds and investment bankers. That's not faith inspiring when looking for an outside CEO to turn around a semiconductor company.

This is why I think the Intel BoD (Board of Director's) are the LARGEST problem.

They won't fire/quit themselves.

They hire those who think short term to: (Make Stock # Go Up) and not think long term.

Any rough patches, they find a "Fall Guy" and fire them like Pat Gelsinger.

Look at their hiring history:

Before Pat Gelsinger, we have:

1) Bob Swan = Bean Counter
2) Brian Krzanich = Mr. DEI & Politics man
3) Paul Otellini = Mr. MBA who down-sized the work-force by 10%

Craig Barrett might have been the last good CEO before a string of 3x Bad CEO's.

---

I will defer to my cohort Paul on this one, as I don't really know much about various CEOs / executives. Here's a convo regarding Lip-Bu:
-------------
Lip-Bu Tan could be the savior, dude knows how to run a fab. He is royalty in the chip world. The only person more revered is Morris Chang. He understands the way the fab networks work.

He has been involved in all kinds of startups and chip companies spanning decades. He gets it. He is an industry guy. So the rumor is that Lip-Bu Tan left the board to run a committee studying how, when, and if Intel should spin off the fabs. He is a big proponent of that, at least according to said rumors, and had disagreements with Gelsinger about that strategy.

So, if Lip-Bu is in fact their pick, one could assume they would plan to spin off fabs. As CEO of Cadence, he interacted with the entire third-party foundry industry. Like, he knows the people, he has connections, he knows how to do things. He would be a superb pick.

https://www.imec-int.com/en/press/jim-morgan-and-lip-bu-tan-receive-imecs-lifetime-innovation-award
-------------

And my personal take? Even someone with all the right connections and background may not be able to succeed, given Gelsinger failed. And spinning off the fabs is still messy business, could impact CHIPS Act funding, etc. Even after AMD split off GloFo, it was years before the albatross around AMD's neck due to prior contracts went away, and the same would likely happen with Intel.

But since Intel signed the CHIPS act, Intel would still need to own 50.1% of any Spin-Off Fab.

So basically, the IDM (Integrated Device Manufacturer) model has to stay because they signed on the dotted line.

Even if they seperate IFS into a Pay for Play Foundary, it'll still be under the Intel Holding Parent Companies purview.

So Lip-Bu Tan will have a very interesting hand / situation given to him to solve with lots of restrictions thanks to the acceptance of the CHIPS Act.

 

[jkflipflop98]

This is the worst outcome possible. Pat's plan was solid. We knew it was going to get worse before the big payoff - and that payoff is just around the corner. He made all the right moves to set the company up for the future.

Now we're going to have some bean-counter idiot in charge that can only think 1-2 quarters ahead and sees nothing but dollar signs. This whole thing just blows my mind.

 

[hotaru251]

worst part of this is Gelsinger was tossed into a mess he never had any actual chance of turning around in his few yrs return.

The replacement will be in a much better spot even if investors only see short term issues.

Would be funny if this move actually is what makes em crash even harder

 

[Mama Changa]

Seeing how screwed up America is currently, I vote for fElon Musk to be CEO of Intel and stage a hostile takeover of AMD.

 

[phead128]

For all those saying "he didn't have enough time", Pat Gelsinger made the bet of "5 nodes in 4 years", and he served for 4 years (3 years, 10 mons). 20A got cancelled/outsourced, 18A is terrible shape. The Board generously gave him 4 years, and his bet failed. Who set "5 nodes in 4 years" must meet their own timelines.

 

[shady28]

I will defer to my cohort Paul on this one, as I don't really know much about various CEOs / executives. Here's a convo regarding Lip-Bu:
-------------
Lip-Bu Tan could be the savior, dude knows how to run a fab. He is royalty in the chip world. The only person more revered is Morris Chang. He understands the way the fab networks work.

He has been involved in all kinds of startups and chip companies spanning decades. He gets it. He is an industry guy. So the rumor is that Lip-Bu Tan left the board to run a committee studying how, when, and if Intel should spin off the fabs. He is a big proponent of that, at least according to said rumors, and had disagreements with Gelsinger about that strategy.

So, if Lip-Bu is in fact their pick, one could assume they would plan to spin off fabs. As CEO of Cadence, he interacted with the entire third-party foundry industry. Like, he knows the people, he has connections, he knows how to do things. He would be a superb pick.

https://www.imec-int.com/en/press/jim-morgan-and-lip-bu-tan-receive-imecs-lifetime-innovation-award
-------------

And my personal take? Even someone with all the right connections and background may not be able to succeed, given Gelsinger failed. And spinning off the fabs is still messy business, could impact CHIPS Act funding, etc. Even after AMD split off GloFo, it was years before the albatross around AMD's neck due to prior contracts went away, and the same would likely happen with Intel.

Here's the thing about Lip-Bu. If you read on when he left, the main issue he had was that the company was full of complacency and was risk averse.

So if it's Lip-Bu, then that would imply that Gelsinger was not sufficiently aggressive or enough of a risk taker, and had not altered the corporate culture rapidly enough.

I find that to be an interesting thought.

 

[8086]

ah right, getting a investment focused person over engineer personale is one of the dumbest move move for a tech company where it took years to let tech mature

This was the problem GM and Ford had in the 70s and 80s.

 

[JRStern]

The Board gave him 4 years (Pat served 3 years 10 months) to pursue his aggressive "5 nodes in 4 years" plan. We saw 20A got cancelled, 18A is only 10% yield. The original plan is not 5 nodes in 6-7 years, at an unsustainable burn-rate of $17 billion per quarter! Let's Pat be accountable to his own timeline.

About that 10% - what are they expecting at this point?
I strongly suspect TSM's leading edge nodes are doing very weak numbers too - and all EUV nodes will always do awful yield.
Are there any real numbers out there for anyone doing EUV?
(or even the last DUV nodes, dual patterning, quad patterning ...)

 

[ottonis]

Cadence Solutions is a company that is involved with Chip-design and AI solutions. Mr Tan is not just a random "investor" but has a degree in physics and a master in nuclear engineering, alongside a master of business adminstration.
This alone would make Mr. Tan a candidate worthy of being considered as CEO of Intel, as he has skills and experience in many of the core areas related to a chip-developer and manufacturer such as Intel.

The problem, however, is not the CEO. Pat Gelsinger did a great job in my humble opinion but was brought down by a board that is apparently not patient enough for major restructuring processes. What their board wants is: Intel should teansform its business AND Intel should accomplish that within 3 months and without any costs or losses.
This is a feat that not even a magician would be able to pull.
So, what needs a change is not so much the CEO position but rather the Intel executive and advisory boards who apparently consist of people who value their personal bonuses more than the fate of the company.

 

[phead128]

About that 10% - what are they expecting at this point?

Pat Gelsinger claimed a defect density of 0.4 (source), which would translate to >70% yield depending on chiplet size and yield curves towards HVM. The problem is Intel is stuck at 10% with HVM in 2-3 quarters, it's not viable for HVM, and no progress made.

I strongly suspect TSM's leading edge nodes are doing very weak numbers too - and all EUV nodes will always do awful yield.
Are there any real numbers out there for anyone doing EUV?
(or even the last DUV nodes, dual patterning, quad patterning ...)

TSMC yields are 70% at N3 (3nm) (source) and Samsung's 50-60% for 3nm (1st Gen SF3E) . (source). In contrast, Intel has 10% yield at 3nm (18A) is truly stunning and resulted in Pat Gelsinger's departure. For reference, it's alledged that China's SMIC has <50% yield using DUV quad patterning for 7nm (reference, TSMC could do 60-70% for 7nm DUV only, and 90% for 7nm EUV), so Intel is on another level is poor execution at 10% using EUV. This is a major reason why Pat was fired.

TSMC yields are 70% at 2nm (source), and boosted extra 6% yield improving just few days ago (source), which translated to tens of billions in cost savings. In comparison, Samsung yield at 10-20% at 2nm (rebranded 2nd-gen 3nm) .(source) which is not viable for HVM.

What their board wants is: Intel should teansform its business AND Intel should accomplish that within 3 months and without any costs or losses.

What the board wanted is Pat to meet his own timeline of "5 nodes in 4 years", and after 4 years on the job (3 years, 10 mons), by all measures, his risky ambitious bet failed. It's not the Board who promised "5 nodes in 4 years", Pat has to be accountable to his timelines.

 

[magbarn]

and we have seen it before cough cough Brian Krzanich

And countless poor examples of bean counters/marketing type CEO's ruining companies like Pre 2nd coming of Job's Apple, GE, and Boeing. These guys are the parasites of the tech/business world.

 

[ottonis]

What the board wanted is Pat to meet his own timeline of "5 nodes in 4 years", and after 4 years on the job (3 years, 10 mons), by all measures, his risky ambitious bet failed. It's not the Board who promised "5 nodes in 4 years", Pat has to be accountable to his timelines.

Are you sure that this was actually written in his contract and not just a mere marketing statement? The thing is, that a CEO absolutely cannot even remotely make such promises, considering Intel's more than half a decade long struggles with its 10 nm node.