[0:03]We are at a crossroads. This is the moment in history where people will look back and say CIOS and AI leaders either made a mistake or they helped put us on a path to greatness. Right now, there are two equal and opposite errors that we can make with AI. The first is to disbelieve in AI's transformative impact because we haven't seen enough of it. The other is to believe it too much, to fall for the hype and visions of magic. Somewhere in between those two extremes lies greatness, the golden path where value can be found. Today, we will give you a road map to walk that path and to find that value. Welcome to IT Symposium Xpo.
[0:58]I'm Alicia Mullery. And I'm Daryl Plummer. AI has shown us some amazing possibilities, but when it comes to AI value, there is cause for concern. In 2025, Gartner found the odds of an AI initiative achieving ROI were one in five. And the odds of an AI initiative achieving true transformation were one in 50. As AI slides from the peak of inflated expectations towards the trough of disillusionment, being trusted is critical. Good news. Gartner's latest C suite survey finds that you CIOS are the second most trusted C suite member in high growth companies, behind only the CFO. This is your hero moment. But heroes are not made at the peak of hype. Heroes are made in the trough. CFOs have told us that 74% of you are already getting productivity gains. However, only 11% of organizations see clear ROI value. Be a hero. Take your AI use cases to the next level. Because the road to value is not paved with productivity wins alone. And value? Well that's in the eye of the beholder. In the private sector, value is growth. In the public sector, it's mission success, and all of you want cost reductions. But is AI ready? Are you ready? Readiness is the key that unlocks that next level, and here's the headline. While not all AI is ready to deliver value, humans are even less ready to capture value. AI readiness doesn't mean can it pass the world's toughest test or can it outthink a human? It means, can it help you find value? A lot of what's going on today isn't even about value. It's just AI feeding on itself. I'm going to let you say that. Enshittification. Thank you. Human readiness is about whether you have the right workforce and organization to capture and sustain AI value. 87% of employees are interested in using AI tools, but only 32% are confident in leadership to drive AI transformation. So take a look at this. In the beginning, AI and humans are not ready, and this is where you should be highly skeptical of AI. But when we cross over a threshold where both are ready, we flip a switch into justified AI optimism. The problem is, you are here. Almost halfway up AI readiness, but only one quarter of the way along human readiness. Value is hard to achieve when you're so far out of balance. AI readiness grows much faster than human readiness. If all the vendors stopped innovating with AI today, we'd still have years before we could catch up. That's why it feels like we can't keep up. For you to get value, you need to know where in this picture your use case falls. Think of each use case as a step on the golden path. This is your you are here GPS. It helps chart your course on the path to value. Wait a minute. That's better, the Gartner positioning system. Can only get it here. Think of this is like navigating a map. Following the GPS, you'll seek to find, capture and sustain AI value. And if you're successful, you can transcend your limitations. But before that, when AI and human readiness are both low, value is in defending your position, and that's primarily about augmenting employees. Now, this makes up nearly half of all AI investments, and the problem here is misplaced expectations. Your organization is expecting one type of value, but investing in another. When AI readiness is low, it's difficult to see what's ahead. It's like a fog has rolled in and you can no longer see the path. But if you can navigate it, you'll find new sources of AI value. Now, someone out there is thinking, come on, how can you keep saying AI is not ready? People are getting lots of value from AI. And that is true. We've been using traditional AI techniques like neural networks and reinforcement learning for decades. But Gen AI technical capabilities, costs and vendors are all of questionable readiness. Let's start with technical capabilities. Now, things such as search, collating data from multiple sources, content generation and summarization, we do those things with AI all day, every day. So let's call those ready. But at this point in the path, to find value, two points of interest should concern you. AI accuracy and AI agents. Gen AI has an error rate of up to 25%, depending on the use case. Now yours might be a company that can live with that. But what if you're the Federal Reserve Bank and require a 0.001% error rate in payments? Gartner finds that 84% of CIOs and IT leaders don't have a formal process to track AI accuracy. In fact, the top approach used today is human review. But the human in the loop equation is collapsing on itself. AI can make mistakes faster than we humans can catch them. Or worse yet, facts themselves can be distorted. Gartner has just published a book on this called World Without Truth. Sometimes you need a set of options, but sometimes you just need the right answer every time. Like when I asked Bjorn, my partner, how do I look? There is only one right answer. You need to bring your own accuracy. Let's call it your accuracy survival kit. In it, you'll find formal metrics, which IT uses, like a comparison metric where you test AI output against an established norm. Two factor error checking, which every employee should too, where you get one AI model to check another. And the good enough ratio. This is a measure of when AI accuracy is just good enough for your initiative. And by the way, good enough may be harder to achieve than you think, because we hold AI to a higher standard. Another must-see point of interest, AI agents are at the top of the hype cycle. Currently, 17% of CIOs report that their organization has already adopted AI agents, and an additional 42% are planning to adopt within the next 12 months. But not all agents are created equal. Globally, 88% of IT leaders are focused on conversational agents, and conversational agents are ready to have conversations. But if you need them to make decisions, and you should, then conversational agents are not ready. Decisions and reasoning are necessary for autonomous multi-agent systems. Remember the first time you called a customer support line, and they kept going through a script, no matter what you said, they just kept repeating the same thing? Did you restart? Is your monitor on? That's what it feels like when agents can't make decisions. Using agents to handle conversations is missing the point. For example, retailers are not looking to conversational agents as the next big AI when. They need something more ambitious, more ambitious than automatically ordering clothes online, or even doing a customer support call. They need agents that can watch what customers are buying, trigger multiple RFPs for replenishment, negotiate terms and conditions and decide the best offer for their company. That is multiplex B2B negotiation, and conversational agents don't do that. Don't settle for conversations. Demand reasoning and autonomous decision making.
[9:37]And AI agents should handle specific jobs to be done. For me, I desperately wanted agent that builds my presentation for me while I talk to it. That's an agent acting as an expert in graphic design. And what if that agent was an expert that could do building approvals, or do my taxes, or the company's taxes? As AI agents as experts can create real world value, they'll do it as long as you're clear on what you need from the technology. And these technologies aren't cheap. So we have to consider the true costs. CFOs tell us, they are losing track of what is being spent on AI projects. They knew the cost on day one, they don't know the cost on day 100. 74% of organizations are barely breaking even or even losing money on their AI investments. But if previous tech introductions had upfront transition costs, AI has a transition mortgage that you just keep paying. Why? Well, when you implemented ERP, you licensed the software and trained your people on one kind of work. But with AI, the work and the training keeps switching context. First you learn to summarize emails, then it helps you do data analysis, and after that you're creating compelling videos. You're using one model before lunch and a different one after. You need people to have experiential knowledge, connect them with peers and show them what good looks like. Because most people are just discovering what they can do, not being trained on what they should do. Here's the bill. AI implementation costs, average 1.9 million per company, and that's just the bill on day one. But then you need to invest heavily in AI training days and literacy programs, because for every 100 days of implementation, AI adds 25 more days to train staff. And worst yet, change management costs add another 100 to 200 days. That's up to 200% more effort. Next on the bill, is ancillary costs. For every AI tool that you buy, there will be 10 hidden costs you didn't anticipate. Costs such as managing access credentials for autonomous agents, acquiring new data sets to ground AI, or packing your accuracy survival kit. There's a hidden transition cost to all of that. So what do you do? For every AI tool you buy, anticipate 10 hidden costs. Conduct an analysis and decide which costs you'll fund. The last thing you want is to be the unintended owner of a negative ROI business case. And you still need to select the right AI vendors for your needs. Now, if choosing an ERP vendor was like getting married, then choosing an AI vendor is more like getting married, having triplets and moving to a different country. It's complicated. Just like kids who touch everything with their little hands, AI is touching everything. And here's why. When marrying an AI vendor, their models are like your children. Children that need to be fed the right data to grow, constantly re-educated and grounded. And those children, well they take on your ethics, and the ethics of the vendor you're married to. And it's not just about the kids. You have to pick the right country to move to. In the AI race, those vendors are starting to resemble countries to live in. We call them Digital Nation States. A digital nation state is a vendor that controls land, power, water, talent, and capital to rival those of actual nations. Large vendors spend more on AI infrastructure per quarter than the individual, annual GDP of more than 47% of the world's countries. If you're planning a massive rollout of AI to your enterprise, bet on the major hyperscalers. These are players like Microsoft, Google, Amazon, Alibaba, and Oracle. They are superpowers, sometimes with government support and always with massive ecosystems and AI infrastructure scale. If you want industry specific use cases, partner with startups and industry leaders. The combination of innovation and domain knowledge comes with access to new markets, customers, and data, and this is more like a developing country. Bet on wildcard vendors such as OpenAI, Anthropic, Meta, Deepseek, and Mistral AI. That's if you want leading edge capabilities. They don't have the scale and reach of a superpower yet, but their influence is rapidly growing. You should note, these are newly minted nation states, and their behavior can shake the ground under everyone else. They're fully innovation ready, but not fully enterprise ready. As a potential AI bubble builds, there are a lot of circular deals being made with digital nation states. For vendors, these deals can be great because they create money seemingly out of thin air. But with all the heavy investments, some will never recover that money in real revenue. And if the attention that they have is so focused on AI, how much attention is left to focus on the products that you already own? So don't select a vendor mostly based on its high valuation or just its AI technology. Select it based on its ecosystem of partners and their shared ability to deliver game-changing solutions or outcomes to you. That finds value regardless of whether there is a bubble or not. Picking a vendor to marry is hard, so here's an easy button for you. If you're going to select an AI vendor today, you had better select one that is good with AI agents. Gartner created an Agentic compass that matches vendors with your requirements. It learns your use case characteristics, maps them to AI capabilities, and then points you to the AI vendors who provide them. We'll put vendor selection on the map here. But bear with me a moment, because there's one more thing about vendor selection that should be important to everyone here, and that is AI sovereignty. Now, most of you are used to dealing with data sovereignty. But with AI, a lot of data is hidden behind the models. To ensure sovereignty, you can't just protect the data. How do you even know what sovereign data has been trained into a model? You have to protect the model and the results the model generates, and who has access to those results. And guess what? You can get locked into a model or locked out of a relationship. The US, China, and the European Union are building sovereign AI solutions with proprietary models that they control. By 2027, 35% of countries will be locked into region-specific AI platforms using proprietary contextual data. Now if you're a global enterprise, that could be very disruptive. And even if you're not a global enterprise, you might get locked out of a long-term partner relationship. So here's what you do. Build, buy or steal, don't steal anything. Acquire a digital tokenization solution to anonymize data so the real data doesn't leave your shores, even inside a model. Like EWS, who owns the payment network Zelle, they are embedding Capital One's Data Bolt tokenization platform to protect data and transactions. Data Bolt handles anonymizing and de-anonymizing the data for you, so you don't have to manage a bunch of encryption keys. And consider open source models to potentially avoid model lock-in altogether. Looking over our map for AI readiness, we can see we're making progress. AI accuracy is not ready. So bring your AI accuracy survival kit, pivot from conversational agents to decision-making agents, even if it forces you to switch vendors, and invest in AI agents as experts. The day one AI bill is ready, the day 100 AI bill is not ready. Anticipate the ancillary costs of moving to AI, costs such as managing agents and multiple models. Don't select vendors solely on technology or valuation, use Gartner's Agentic compass to select the right vendors and realize that every AI vendor selection is a sovereignty decision. So track the rise of digital nation states. We've just gone through the fog of finding value, which answers the question, can this be done? When human readiness is low, we must ask ourselves if we can capture and sustain value. Not just can we do it, but can we keep doing it? The golden path here is littered with obstacles, which will slow you down. Our people alone can be a big obstacle. I live in London, and every Sunday, Bjorn and I love to walk to the farmer's market. I just want a nice stroll, but he is always in a rush to get his favorite pastry, and this is what low human readiness feels like. It's like, come on, Alicia, hurry up already. If human readiness is low, it will slow you down. In fact, 71% of CIOs and IT leaders report their workforce is not ready for AI. Why? Because AI unleashes a toxic mix of a steep learning curve and the primal fear that AI is going to replace us. Headlines everywhere claiming that AI is taking jobs and replacing staff. The Anthropic CEO says AI could wipe out half of all entry-level white collar jobs in the next five years. And 60% of CFOs are expecting they can reduce head count due to AI. Those kinds of statements make it seem like there's an AI jobs bloodbath going on. But is that true? Nope. We find that only 1% of head count reductions are directly due to AI today. Let me say that again. Only 1%. It's not about job loss. It's about job chaos. And it's your job to settle the chaos. In fact, job and roll redesign is a 20 times bigger effort than layoffs or hiring. Now, yes, people will lose jobs, and that matters to us all, but it's not a bloodbath. What we do see is hiring restraint for junior-level jobs. Senior staff can now delegate to AI much of the work that juniors used to do. That captures value. But the problem is, we keep getting distracted by headlines. You may have read news recently about Salesforce eliminating 4,000 jobs. What didn't make the headline was that while they are cutting customer support staff, they are also hiring staff to support net new AI revenue streams. This is a talent remix, and it seems like all the big tech providers are doing it. Shifting talent from lower performing business units to higher potential AI driven business lines. Unless you're a tech company, this strategy isn't your strategy. Your ability to set up whole new lines of business is limited, so your returns from copying this strategy will be limited. Instead, your strategy needs to be a value remix. Look at how AI can cut your backlog, how it can reduce fraud, or how it can grow revenue through human empathy. Like Sanlam, the South African finance company, whose customers ask for loans to pay off a rising debt. Sanlam uses AI to empathize with customers, and help them accept the idea of a debt reduction plan, instead of a new loan. The value captured here is not from head count reduction, it's from less bad debt and healthier customers. And guess what? All of you have a human empathy opportunity somewhere in your organizations. But almost none of us are thinking about it as our next big AI breakthrough. Tell your CFO that you need a value remix, not a talent remix.
