Thread regarding IBM layoffs

Do analysts ever get demoted or fired for their mistaken predictions?

Is IBM the Next Hot AI Stock? This Analyst Says Watson X Software Is the Company’s Most Promising New Product In Years. --

https://www.barrons.com/articles/ibm-stock-price-ai-watson-x-bc2a4ab1

By: Eric J. Savitz
July 20, 2023 12:51 pm ET

The next hot artificial intelligence stock just might be IBM .

The enterprise computing giant posted better-than-expected financial results for the second quarter, driven largely by strength in its software business. On the company’s earnings call Wednesday, IBM (ticker: IBM) CEO Arvind Krishna spent considerable time talking about Watson X, IBM’s new software platform for building enterprise AI models and applications. And the Street is impressed.

“Watson X is the most promising new, organically developed software product IBM has launched in years,” MoffettNathanson analyst Lisa Ellis wrote in a research note following the earnings report. She thinks Watson X could be “a new blockbuster for IBM,” exceeding Red Hat’s hybrid cloud platform.

Krishna said that in just the first 10 days of availability, Watson X software has been deployed to more than 150 clients, among them Samsung, Citi, and NASA. As Ellis notes, IBM has no plans to offer a general-purpose chatbot like ChatGPT. Instead, IBM is focused on meeting enterprise AI requirements, including data privacy, integrity, and governance.

IBM is no newcomer to AI: Big Blue has been working on artificial intelligence software for longer than almost any other company on Earth. You might recall that IBM’s DeepBlue chess playing supercomputer beat the world champion Garry Kasparov in a 6-game match in 1997, more than a quarter-century ago.

And yet, while the market this year has dived headlong into almost every AI-related stock imaginable over the first half of the year, IBM has largely been left out of the party: Shares are down 1% so far this year. Compare that to Nvidia (NVDA), up 214%, Microsoft (MSFT), up 45%, and Alphabet (GOOGL), up 36%.

As Barron’s has reported previously, Watson X has three parts: Watson.ai works with customers to create new models, or data sets. Watson.data acts as a data store, putting the company in competition with Snowflake (SNOW), among others. And Watson.governance monitors AI models to make sure they are accurate and accountable, not filled with false or offensive information.

“There are endless potential use cases for generative AI models in an enterprise context,” Ellis writes, “such as augmenting HR, performing predictive IT maintenance, helping with financial forecasting, performing fraud detection, compliance monitoring, risk management, and supply chain optimization, among others.”

She says she sees potential for generative AI to “transform a wide range of business processes, ranging from industry-specific use cases, such as dr-g discovery, to functional use cases, such as software coding or call center operations.”

Krishna talked extensively about the potential for Watson X on the company’s earnings call. “Large language models are a step change in the evolution of AI, with more than 80% of enterprises exploring their use,” he said. “We believe the opportunity for large language models for enterprises is immense given the existing amount of business data. This includes sensor data, chemistry data, materials data, geospatial data, code, and of course, speech.”

Ellis maintains her Market Perform rating and $145 target price on IBM shares. After posting better-than-expected profits for the second quarter, IBM stock on Thursday is up 3.2%, to $139.75.

Wake up! Gartner is a paid for Marketing wh0re Anyone with true business acumen gives their reports zero credence

If you do a Google search you can find a January 2023 Magic Quadrant chart that shows IBM as a leader in both vision and ability to execute...so they've got that going for them.

Setting aside the problems with Watson Health, this begs the question...what is Watson actually good at? It has great (some might say fantastic) reviews for analytics and natural language querying, so maybe it has a future. However, its closed nature might relegate it to only niche applications (which may be what the management wants).

Has Gartner come out with a magic quadrant yet that includes WatsonX is or it too early yet. A lot of customers pay for Gartner and value their feedback

There have been multiple threads lamenting the lack of innovation at IBM.

watsonx could be a better offering if IBM offered both prompted and programmable access to internal employees to try it out, write programs to talk to it, etc. Kyndryl has been spun out, so a lot of the legal difficulties no longer exist. If they are not doing it, one wonders what would prevent them from doing so.

There is still no info within IBM on how to access their WatsonX prompt lab. Anyone who asks is greeted with silence.

IBM said they were "ALL IN" on Hybrid Cloud.
Now that Generative AI is the buzzword of the moment, does that mean IBM is no longer "ALL IN" on Hybrid Cloud; and is now only half in on Hybrid Cloud, and half in on Generative AI? Or, is IBM now "ALL IN" on generative AI with Watsonx, and now not in at all on Hybrid Cloud?
What will be the next buzzword that IBM will be "ALL IN" on? :)

You would think that watsonx would be accessible by IBM internals and promoted internally. But then again, I remember how hard it was to get IBM software back in the day. If a particular offering was unavailable in public demo or sample form, then divisions outside SWG had to pay an internal license fee to SWG for access to the software. I can only imagine what kind of games an IBM internal has to play these days to get access to the AI stuff.

If IBM wants to limit access to their AI, Watsonx, to paying enterprise clients, IBM should at least allow those that work at IBM to access some kind of generative AI prompt like ChatGPT. But, IBM doesn't even do that. I've looked all over IBM's internal intranet and have found nothing.

IBM can command pricing if they have a monopoly or have a superior offering. Outside of the mainframe market, they have neither one as either B2B or B2C. Their AI offerings could potentially improve if they have more data sets and more ways to test their stuff, but the tight focus on their existing customer strongholds will probably leave them an also-ran. Maybe they'll make a compelling watsonx offering for their customers, but will it really be better than Bard or GPT? We'll see.

IBM is and has always been a B2B company (Enterprise installs based on monopoly or unique advantage (eg OS/400 or MVS)) As soon as competition forces the market to commodity (B2C), IBM has a very difficult time competing due to costs (PC’s, intel servers, UNIX and their servers, storage, services, cloud). Once that happens, IBM is forced to restructure or withdrawal. Better to play where you are the only game in town as you can command pricing

IBM has a rather severe structural problem that will probably prove impossible to solve in the long run. IBM has always been and will always be a "high priest" company in a world that is struggling to get rid of high priests.

Amazon, Google and Microsoft made their success in cloud computing by making stuff that was accessible by the masses. All AWS, Google Cloud and Azure services are easy to buy (use a credit card), and it is "easy" for a skilled developer to make stuff that works on those platforms. The documentation is available to everybody, testing and development is relatively straightforward, and once you have an offering each platform has a published means of deploying said offering to everybody else.

In AI, IBM is once again limiting its offering to enterprises...in other words, customers that offer the promise of easy money. It's what they've always done. You want access to the watsonx container libraries? It's just like those IBM Cloud Paks...you need to go through Passport Advantage...good luck with that if you're just an ordinary individual and not an existing IBM customer. It would be different if IBM had first-mover advantage and something compelling to work with, but OpenAI is already offering free credit for experimenters to work with their offering. You contact sales through their website, and the "OpenAI" brand is already known thanks to GPT.

If I'm an AI developer, then I know who I'll start working with first. MAYBE I'll work with watsonx in the future if the platform proves itself and there is sufficient demand...or maybe I'll just stick with OpenAI and whatever the other players make. At least I know I'll be able to get easy access to their stuff and have a chance of making a decent offering.

Wall Street has very different definitions for cloud depending on who is doing the defining. We can make a couple of observations about cloud that definitely apply to IBM

1. IBM completely missed the scale out cloud boat
2. Amazon and Google have first mover cloud status and thus competitive advantage
3. Microsoft has exploited their Windows monopoly to get cloud competitive advantage
4. IBM had to redefine what they define as cloud due to missing the cloud boat
5. IBM has focused their cloud offering on their “Monopoly” (enterprise) Yep Mainframe
6. All of IBM’s developmental efforts feed the “enterprise” offerings as that’s the strategy
7. IBM continues to reshape the company in an Enterprise image. Everything else has been de-emphasized and management will allow it to fade away
8. IBM’s “new” strategy (hybrid enterprise cloud/SW modernization/consulting) can’t be called a failure or a success as it’s still a work in progress. HW = falling revenue, SW = growing revenue, and consulting = quasi flat revenue.
9. IBM has to get their costs and priorities aligned and execute faster

woke company now a dead company

IBM cloud had a long time stigma of being unstable sad to say. Someone clearly missed the boat on the importance of non-functional requirements.

IBM was very late to the cloud party. IBM put some mediocre engineers in charge of its cloud development efforts. After a decade, nothing to show for. IBM could not make a dent in Amazon, Microsoft and Google dominance.
So, if the past provides a clue, IBM is left in the dust in the AI domain.
IBM executives lack vision, they are no more than con artists.