SAP and Apple are close partners (we use a lot of Apple products and Apple uses SAP Business Technology Platform for innovation among other things) and so our developers had early access to the platform and two SAP applications were already available at the launch (SAP Mobile Start and SAP Analytics Cloud), with more on the way.
SAP has been working with customers on the uses of spatial computing at work for a long time, in areas such as augmented reality for technicians (showing virtual plans and procedures in the air next to the machine to be fixed) and warehouse stock-picking.
But I’ve been skeptical about wider use of these technologies in business, mainly because we’ve been talking about it for a long time and not much has come of it.
And more recently all of the publicity around Meta’s vision of the Metaverse did nothing to change my mind—I have zero interest in attending legless avatar meetings or shopping experiences.
My views changed somewhat last year when I saw a great presentation by Oliver Gutzeit, the leader of SAP’s Global Experience Technology Unit.
Oliver’s presentation, demos, and wide variety of use-cases helped convince me that the technology was maturing to the point where it was becoming really useful. Here’s one of his slides from the AFSUG user conference last year:
The term “Metaverse” seems to have fallen by the wayside, and Apple’s spatial computing term seems to be winning — a recent SAP Labs Talk episode interviewed Oliver on the topic.
A big part of my skepticism has been because of the limits of existing technology and the form factor. Now that I’ve tried the Apple Vision Pro, I think things might be changing faster than I realized.
The Apple Vision Pro was very easy to use: It only took me about 30 seconds to get the hang of the new interface techniques, and the “passthrough” vision of the outside world works amazingly well. I more or less forgot I was wearing it almost as soon as I put it on (but didn’t wear it for very long).
Adoption will start with entertainment and games (the spatial videos look amazing), but as the gear gets cheaper, lighter, more powerful, and more AI-enhanced, I can see why this could be an exciting part of the future of work.
While it was impressive, I don’t think I’d use the device much in its current form, even if I were willing to shell out all that money (although the idea of sitting comfortably on a plane or train and being able to work comfortably on a big virtual screen does have some appeal).
I might change my mind when I get to try out Apple’s new more depth-aware spatial video recordings which apparently give you the immersive feeling of being there. But personally I think I’ll probably be coming to spatial computing from the other side, with regular glasses that have been enhanced with cameras, audio, and AI chat functionality, like the Meta Rayban smart glasses or the new Frame glasses.
In the meantime, the SAP IT and innovation teams (and some lucky very senior executives) will be in the front lines of experimentation, sharing what we’ve learned, and increasing the range of SAP applications available.
And if you do have an Apple Vision Pro (or are thinking about getting one for your own corporate innovation center!) here are some links to more examples and screenshots of what you can do with SAP & Apple Vision Pro, and expect more in the near future!
SAP and Apple Partnership
SAP Apple Vision Pro and SAP BTP
SAP Mobile Start on Apple Vision Pro
Researchers say that AI projects tend to be “technosocial” — where technical aspects dominate the social side of work.
But investments in “sociotechnical” approaches (putting people first) can pay big dividends, because ultimately the biggest benefits of AI are in Augmented Intelligence, empowering people and teams to do more than they could in the past.
Taking care to structure the new ways of working after a technology implementation can save a world of hurt and frustration for everybody involved.
According to the academics, there is a list of things you need to think about when optimizing roles and teams for new technologies like AI:
Joint optimization: People should be viewed as complementary to the machines rather than as an extension of them.
Adaptability/agility: The design of work should aim at increasing variety rather than decreasing it, and groups should have responsibility for solving local problems. Individual and organizational learning is essential to allow organizational adaptation to change.
Responsible autonomy: Work should be given to teams or groups, internally supervised, thereby avoiding “silo thinking” by engaging the entire system.
Wholeness: the objective to be completed should be specified, with a minimum of regulation about how it is to be done. The system should be conceived as a set of activities making up a functioning whole, rather than a collection of individual tasks.
Meaning: Each individual should experience an optimal level of variety, have learning opportunities, a scope for setting decisions that affect the outcomes of work, organizational support, a job worthy of societal recognition, and the potential for a desirable future.
Have you seen people discuss these aspects of work in the projects you’ve been involved with?
For more details, read the full paper on Artificial Intelligence and Digital Work: The Sociotechnical Reversal
]]>Generative AI can help turn data garbage into business gold.
Data remains the biggest and most important factor in the usefulness of AI systems. Algorithms are becoming a commodity, so the biggest differentiator is the quantity, quality, and relevance of the underlying data set. And the better the data, the easier it is to create quality outputs.
But there’s an important distinction between the underlying data and the way it’s actually recorded and stored. Real-world systems see the world through a cracked and smudged lens. But even if each point of light is dubious, we can still get an overall impression of what’s going on.
For example, if your IoT sensors are recording random numbers, you obviously can’t get anything useful out of them. But if they’re “just” inaccurate, with the real data hidden behind a veil of noise, the result is still potentially useable with the right statistical techniques. Machine learning algorithms can capture the underlying patterns that (probably) generated the observed, messy data.
Now new Generative AI technologies are providing another huge step forward in dealing with imperfect data.
Large language models are very good at dealing with some types of messy data. For example, researchers have shown that large language models like GTP-4 can decipher even very scrambled sentences:
A personal example: my daughter recorded a short section of her economics class (with permission). The quality was awful—the teacher’s voice was almost completely drowned out by the sound of my daughter typing and other background sounds. I personally couldn’t really hear what he was saying.
I ran the recording through OpenAI’s open-source transcription algorithm Whisper, using the slowest and most sophisticated model available. It did a good job of deciphering many of the spoken words, but there were gaps, a few words that were clearly incorrect, and the result was hard to follow (the teacher had a tendency to digress and circle back).
I took the transcript and put it into ChatGPT 4, asking it to “take the text and put it into sentences”. As if by magic, out popped a restructured, clear, three-paragraph summary of the economic points the teacher had discussed. It wasn’t what he said, but it was a lot closer to what he meant.
Large language models are good at figuring out what we meant, and the principle applies to many real-world data problems.
For example, machine learning is already used to extract information from documents such as invoices: the date, amount, supplier ID etc. But these models require lots of training data, and don’t generalize very well— if you try to use them against a new layout of invoice that the model hasn’t seen before, then it may get stumped. By adding generative AI, the system gets much more effective at dealing with edge cases and novel layouts.
There are dangers, because these models are designed to synthesize what “should” or “could” be there, no just analyzing what is actually there. From the previous examples, the result may be thoughts the economics teacher never mentioned, or a supplier ID even if one is not included in the document.
Figuring out how to avoid such “hallucinations” is currently the leading edge of AI research—with approaches that include asking the model to double-check itself, averaging out the results of several instances of the model, or an extra check from a dedicated verification model acting independently.
But overall, generative AI is a great new opportunity to open up more data in new ways, to rethink what data sources are available, how they can be used to improve processes—and to turn what looks like data garbage into business gold.
]]>At dinner the other day, my daughter was explaining a complex probability question in a recent mathematics test. My wife asked if she had the test paper, and suggested she put into ChatGPT to get help.
I’m sure everybody can relate with that idea. After all, the entire history of computing has told us that computers are good at maths and logic. So when we hear that artificial intelligence has had a considerable breakthrough we all automatically assume that it’s in the same direction.
But that’s what’s so confounding about the latest large language models. They’re generally terrible at mathematics and struggle with logic problems that children can easily manage.
Instead, what they’re really good at is something we think of as much more human: creativity. They are deeply untrustworthy when it comes to facts and logic, but great at brainstorming and new ideas.
This wave of computing and AI is different. Throw out your mental model of software that does what you tell it to do. Instead, you have to talk to these models as if you’re trying to communicate to an alien species. They respond better to coaxing and threats than they do to orders and commands. Studies have shown that they give better answers if you are polite, promise to tip, or if you say that your life depends on it. It may even get sluggish just because it’s December!
The good news is that the next wave will apparently be back in line with our initial expectations. Researchers are working on models that are dramatically better at step-by-step reasoning, and can work their way through complex mathematics problems like my daughter’s probability test (e.g. the rumors swirling around OpenAI’s Q* or the student math example from Google’s Gemini Ultra).
That really will open up the opportunity space for how useful these technologies are in the real world, and get us a lot closer to the holy grail of artificial general intelligence (AGI).
There’s always lots of debate about whether computers are “really” intelligent and can “actually” think. But as Alan Turning said a long time ago, that’s a question that is “too meaningless to deserve discussion”.
I personally think humanity has too much pride to ever admit that computers are intelligent, even as computers now regularly beat us on almost every standard we can think of for measuring it.
But I think we’ve already passed a real watershed towards AGI. Getting the most out of computer now requires treating them as if they were intelligent—and flawed, just like our human colleagues. And from now on, they’ll just get steadily smarter.
]]>One of my maxims has always been “if the reality is better than the marketing, you’re doing marketing wrong”—but there are limits, and the apparently-somewhat-faked launch video of the new Gemini Ultra model seems to have gone beyond them, generating significant backlash:
“Google admits Gemini AI Demo was at Least Partially Faked: This is just embarrassing” https://lnkd.in/euamvWyS
To be clear, this does seems to be a very capable model, the first to really challenge the complete dominance of OpenAI’s GPT-4. And I’ve heard people argue that this is just a question of timing — for example, taking a video and extracting single frames from it is known technology, so maybe Google just felt that what the video depicts in realtime is an acceptable, short-term extrapolation of what the model can do.
Gemini Ultra’s multi-modal approach, combining text, sound, and image into a single model (rather than suboptimally bolted together like GPT-4 and DALL-E) should be very powerful.
But most of us will not be able to test it until next year. In the meantime the less capable Gemini Pro model is now used in Bard, but that seems to be equivalent to ChatGPT-3.5 (and so a long way behind v4)
In many ways, this was always Google’s race to lose, because of their incredible access to the world’s data. I expect them to be one of the clear leaders in the space going forward.
But their biggest challenge, and the reason they were probably late to the game despite being a machine learning pioneer, is their business model: these technologies could dramatically undermine the advertising business that they rely on for >80% of their revenue.
What do you think?
It was one of my favorite events of the year last week — the annual conference for the members of the UK and Ireland SAP user’s group, UKISUG Connect 2023, in the ICC in Birmingham.
As is customary, Paul Cooper, UKISUG Chairman kicking things off with the results of a member survey highlighting current concerns:
….Take the issue of indirect licensing […] after a bumpy ride, we, along with the other global user groups, managed to get SAP to recognize that the customers needed a more transparent and fairer approach to licensing.
The point is that the road has always been bumpy. But when we work constructively together, challenges can be solved. Right now we’re at a point where we are facing a breakdown in trust. On premise and hosted customers are unhappy, to put it mildly. With the statement from Christian Klein stating that SAP would only be releasing new innovations into the cloud.
This isn’t a small minority of customers. When this was announced, we decided to survey our members. First, before we joined and commented on the debate, research we have conducted amongst our members, shows 79% of you that have moved to S/4HANA are an on-premise or hosted deployment. And that of those that are planning to move to S/4HANA, 70% plan to move to on-premise or hosted versions.
Back in 2020, SAP board member Thomas Saueressig, talked about giving customers choice between cloud and on-premise deployments. We were told that SAP believed in a hybrid future. Yet Mr. Saueressig didn’t once say that cloud deployments came with full innovation while other flavors of S/4HANA didn’t. We were also told that S/4HANA would be supported until 2040. There was no caveats that major innovation could only be delivered to public and private cloud customers, using RISE with SAP or GROW. Why did SAP not tell us that new innovations only going to be available to cloud customers? Did they not think it was important in our decision-making process? These are the questions our members have been asking us.
I’ve heard the more cynical amongst you suggest that SAP didn’t disclose this strategy before because customers choosing to stay on-premise might not have made the S/4HANA move at all. Also there are members that SAP itself have told to remain on-premise. Why? Because the size, complexity and criticality of their SAP implementations were deemed too difficult and risky to move to the cloud. Will they be exempt from this future SAP innovations? If so, surely there should have been told when they were initially investing in their move to S/4HANA.
The bottom line is customers that have moved to on-premise or hosted implementations of S/4HANA feel misled. From a UKISUG perspective, this is a significant percentage of our membership. If SAP refuses to communicate more on this matter, there is a real risk many of those planning to migrate to S/4HANA may change their plans. Of those who have already migrated to an innovation-less future, will they choose to invest in other SAP platforms. I’ve already heard a number of on-premise and hosted customers suggest they expect large-scale reductions in support maintenance costs as they are now running an inferior version of S/4HANA. I can see that point. Surely a less future rich product costs less to support than maintain. Others are asking if SAP intends to refund or partly refund them, is they are not providing the product or service they were told about.
We’ve raised these issues with SAP and today they’ve remained publicly silent and for those of you at SAP listening today, I urge you to engage and work towards a solution that rebuilds trust and address the concerns of our membership, the majority of whom, as we said before earlier, have on-premise or hosted deployments. Work with us to build bridges and enable on-premise customers to access innovation.
In private, some SAP executives have told individual on-premise members that there’s no technical reason why they couldn’t access the majority of innovations through BTP. If this is true, then tell us if it isn’t please stop miscommunicating. As customers, what we want is open and transparent relationships with our vendors like SAP, where we can feel valued and we can trust SAP.
This isn’t something that can be resolved by the local team in the UK island where our collaboration is extremely strong. Instead, it requires action from those at the top of SAP. Rather than choosing to please investors over customers, think about pleasing your customers, as without them, your investors will soon disappear.
At the Germans user group conference in September, SAP’s CEO said no customer would be left behind. Hopefully in coming months we can work together to make this a reality and ensure that all customers and our members have access to innovation and aren’t penalized for investing in an on-premise or hybrid version of S/4HANA.
Before I finish today, I’d like to wish Conor Riordan, who will take over from me as Chairman next year, the very best of luck as he steps up to Chair. It’s been an absolute pleasure to serve as Chair of your user group for the last six years. We’ve continued to flourish and evolve.
SAP UK Managing Director Ryan Poggi was next on stage to respond to the remarks. He emphasized joint SAP and customer success, and the importance of transparency:
The latest innovations are what keep us—and you, because I firmly believe that we live [in the same environment] competitive—in an extremely challenging environment. And what we’ve heard consistently is that you expect us, as a partner, to be one step ahead. And to challenge the status quo when we believe that jointly we’re not maximizing our potential.
Change is never easy and we face the same challenges as everyone. How do we direct a finite number of resources to provide the most value to our customers and ensure that we maintain our place at the forefront of an industry we helped create? —which is what you rightfully expect from us. How do you bring the latest technology, for example, generative AI, to you in environments where they add the most value, many of which require the cloud? And finally, how do we ensure that inflation, which in the last year has been as high as 9% and impacts us all, doesn’t create an environment that detracts from our ability to deliver?
And the reality is, like it is for many of you, this requires some tough investment decisions. But this will never change our commitment to supporting all UKI customers at any stage of your transformation, which is something our board has also committed to. But I also note that no matter how much time we dedicate to this topic on this stage, there will still be questions.
So my commitment to you and my ask of you is that we continue to have these transparent conversations with our teams and encourage a connection to the right stakeholders on both sides. Our door is always open. And as I discussed with Paul when we first met, or met rather, at the podcast, I really do believe that the basis for a true partnership is that open, transparent discussion. And part and parcel of that is what I refer to as understanding the why.
That doesn’t mean that understanding why we make the decisions we make, or you make the decisions you make, will solve everything. But it will, I believe, help create that foundation of trust. Now, after what’s just been said, this may seem a little tone deaf, but I’m sure none of you are surprised to see RISE here or around.
I know that for many of you it has felt like all you’ve heard us say is “RISE is the answer, what’s the question?”. But let me take a moment to explain the context of why we launched RISE. And to everyone thinking, well, to sell more S/4HANA Cloud, give me a moment. RISE was actually born directly in response to customer feedback. You shared with us that it was too difficult to work with SAP, that it felt like SAP was only focused on selling products. You were rightfully demanding more from us. The market was also very clear.
If we want to be the cloud company, we need to be a proactive partner and move from a solution and project-delivery focus to being accountable for the operational delivery and outcome. RISE was born from that feedback, to become the proactive partner that you were demanding, bringing the best of our ecosystem together, while still giving you flexibility of choice in terms of hyperscaler or services provider, all while ensuring SAP remains accountable for the outcome.
RISE isn’t only about S4, it’s a cloud operating model that leverages the power of our ecosystem for you. And this year, we have seen that the understanding of RISE across the UKI has progressed significantly, thanks to the help of our joint workshops and this forum, the user group. We’ll hear a little bit later directly from one of our customers about how and whether they perceive this change.
As I said, understanding the why doesn’t solve everything, but hopefully it provides a better understanding of how we’re trying to build that partnership. And this means evolving the way we partner with all customers, because history has proven that lasting economic turnaround in UKI requires growth in the SME market.
Employing over 16 million people, these organizations make up over 80% of the market. Many of you here today, we cannot overstate the importance of these companies to economic recovery. But not only can SMEs out-compete with the right partner, they’re able to change the entire playing field.
The challenge, however, is not to allow complexity to creep in as they grow. And this is where technology comes to the fore, by allowing you to focus on your growth by providing a platform that delivers stabilization. GROW with SAP makes this accessible to all organizations, a proactive partnership across our ecosystem, giving you the ability to leverage fit to standard in the cloud from day one.
[… my door is] always open and we value transparent conversation. I trust that over the course of this event our team will continue to demonstrate how we are laser focused on earning your trust and the followership in the UKI. I very much look forward to continuing to work with all of you and all of our customers in 2024 and beyond and it’s been a pleasure to be here. Thank you very much.
For more discussion on this, please take a look at this Diginomica article by Derek du Preez
]]>This is a concept that is incredibly relevant to digital innovation—because when the people on stage tell you how to be successful, you’re not seeing all the other people that followed that same advice but failed miserably.
Here, for example, is professor Scott Galloway explaining why the “worst advice you will ever hear in business school” is “follow your passion”
]]>I just listened to industry veteran analysts Jon Reed and Josh Greenbaum discuss the recent ASUG Tech Connect Event in Orlando.
I strongly urge you to take a listen to the whole thing, but one thing jumped out: that bad data is one of the biggest problems organizations face when trying to take advantage of Generative AI.
This isn’t surprising—garbage in, garbage out—but as both Jon and Josh pointed out, this represents a new opportunity to get funding and attention for what organizations should have been doing the whole time.
IT organizations have good excuses: it’s hard to build executive enthusiasm for something as seemingly plumbing-related as data quality. And poor data quality isn’t a technical problem, it’s almost always a reflection of broken business processes and bad incentives—so fixing it without business ownership and support is near impossible.
Generative AI increases the ROI you can get from clean data, and new Business Data Fabric approaches (SAP Datasphere + strategic partnerships) are making it easier to achieve than ever.
So if you want to move forward with Generative AI, you first probably need to take a small step backwards, and build a solid information foundation.
]]>And partners are an essential part of composability, because one of the ironies is that customers shouldn’t have to build anything themselves. The whole point is that other people have already had this problem, built a solution, and you can simply plug and play and put those new Lego bricks into your existing process.
So this is very much part of the move to new business models, where instead of just selling services, you can sell IP, you can sell solutions along with added services as part of your business.
The second big topic is one that everybody is interested in right now is AI. There’s a big fear of missing out of the new opportunities, and there’s a lot to talk about, but technologies like chat GPT have been the fastest adopted technologies ever, with a million users in just five days.
And these generative AI technologies allow new possibilities for creativity. It’s great for things that you can, it would take a long time to build, but you can verify very quickly. So they can help you with coding, for example. It would take a long time to code something. It turns out the chat GPT and Google Copilot and these other technologies allow even non-coders to create code, and then you can very easily check whether it’s actually real or not. Or you can use Midjourney and Dall-E to create images. You just put in some text, it creates an image. You can say, yes, that’s the image I want or not. And they’re very much iterative.
Overall, the net result is that part of all of our jobs and our organizations, the value of what they’re doing is going to be much less than it was before. So we have to be clear about what those areas might be. The good news is that there are other areas of our business, perhaps 10%, where we can do 100 times more by using these powerful technologies.
So it’s a moment to take a big step back, recalibrate, rethink what is that we’re really good at that we can leverage AI. And one of that will be about what it means to be human. One of my favorite quotes in technology is by Pablo Picasso. And it’s “Computers are useless. They can only give you answers.”
They can’t create questions. You, as a human being, can do that. You know what’s wrong. You know what needs to be fixed. And you have suggestions on how to do it. AI, ultimately, will only be a tool. It’s a very powerful tool that we can all use to create more value.
And the key is that AI on its own is useless. It takes high quality data. And most of the highest quality data in most organizations is in their SAP systems. So help get the SAP data into machine learning models, and then, on the other end, actually make sure that something happens. An algorithm can’t do anything. It has to be part of a business process, and that brings us back to that composability.
The next area is data. There’s a huge disruption around data. For a long time, we’ve always ripped data out of our core systems and put it into a data warehouse or a data lake or a data lake house or a data cloud.
Increasingly now, we can bring the technology to the data rather than the other way around. For too long, it’s been like ripping a tree out of a forest and then trying to get it to grow in a different environment. It works, but it’s a lot of hard work. You lose the roots, the metadata, the hierarchies, the real-level security. And then you have to recreate it all in this new area.
We’d like to help our customers avoid doing that with technologies like DataSphere, where people can get more value out of their data in SAP by leaving it as much as possible where it is, but providing services to other areas as needed.
Finally and I think perhaps the most important area, is people power. We have an amazing opportunity to accelerate innovation by letting business people do more of it themselves, in their area of expertise, without IT and technology being a bottleneck.
It’s the rise of the purple people. Lot of technology is now being done outside of IT, so you’ve got the blue business professionals and the red technology professionals. The purple people are what Gartner calls the business technologists in the middle that have a combination of those skills, and they’re responsible for up to 80% of new innovation in the next couple of years, according to Gartner.
Why? Well, it’s thanks to these new no-code and low-code technologies and chat GPT and these others, which means that the power users are now much more powerful.
There are dangers, though. Not everybody is ready to be enabled, and there’s the dangerous chaos where people go off and create their own end-to-end processes, but without any governance, without following the security rules, without compliance such as GDPR, or building things that simply aren’t scalable, or rebuilding the same things at different parts of the organization.
The big opportunity is not pro-code or low-code, but co-code, where you can use the SAP technologies to provide the bricks. It could be a concept like headcount, or revenue, or it could be an action like open a higher hiring position, or send an invoice.
Then you’re going to let the business people manipulate those bricks, connect them up to create those workflows, but you can be confident that they’re doing so in a governed way that isn’t going to put the enterprise at peril.
Those were the top four topics and I really hope at some point I can have the opportunity to explain more!
]]>