Baker Hughes Senior V.P. touts benefits of autonomous well construction

Baker Hughes Senior Vice President, Well Construction, Jim Sessions

Interview with JIM SESSIONS, Senior Vice President for Well Construction, Baker Hughes 

Autonomous drilling has opened new frontiers for the upstream industry, replacing reactive operations with intelligent systems that can learn, adapt and optimize performance in real time. In the words of Baker Hughes, “this digitally driven approach, built on decades of drilling expertise and intelligent engineering, is making well construction smarter, safer and more predictable.”  

Accordingly, Baker Hughes on Jan. 29, 2026, debuted their Kantori™ autonomous well construction service. Kantori™ supports the entire well construction life cycle, from connectivity and data integration to well planning and performance optimization. Recently, the company’s Senior Vice President for Well Construction, Jim Sessions, sat down with World Oil to discuss the overall benefits that autonomous drilling is bringing to the industry, including the unique features that the new service offers to the process. 

Fig. 1. TRU-Steer is the latest advancement from Baker Hughes in the evolution of rotary steerable systems. Image: Baker Hughes.

World Oil (WO): Let’s start with a bit of an overview on autonomous drilling and talk about the specific applications for the new Kantori™ autonomous well construction service from Baker Hughes.  

Jim Sessions (JS): Let me start with what it is, what it does and then we can go into it. When you take a look at what we've done historically, obviously the evolution of drilling, rotary steerables (Fig. 1), and LWD has matured over time. Now, it's part of really putting that combination together with some advanced physics models and A.I. to really make it an integrated, repeatable kind of solution.  

It starts with the rig interface to make sure we're fully connected with the rig into the fluids management system for pumping and cuttings removal, down into the steerable sections, where we can get within a 2-m thickness in a reservoir and obviously optimization for ROP and those types of things for faster well delivery. So, that's the back story.  

When you ask about applications, obviously we're seeing improvements of up to 15 days in shorter well delivery cycles. When you're in a high-cost environment, particularly offshore, you're going to bring your wells on faster. There's also the emissions reduction angle—if you're not going to the rig for 15 more days, then that obviously yields an emissions reduction. I mentioned the tight reservoir navigation ability. When you have some real thin strands and you have to stay in there, it'd be great. Also, if you have water contact or something else you don't want, then you're not constantly adjusting your survey. And if you get to the point that you're going to have some water influx, you’re able to keep water out and only the gas in. That's mainly where it's at.   

Fig. 2. A Baker Hughes SME explains the functioning of the Kantori™ Unified Digital Autonomous Well Construction Solution. Image: Baker Hughes.

The other benefits are obviously if you need to drill faster, to meet any kind of permits or any kind of time constraints. It's not a one-size-fits-all. If we don't have the rig systems that have the systems approaching, you lose some capability. The one thing I'll say about the Kantori™ system (Fig. 2) is because it is a combination of parts, if you don't have all the capability needed to support it, you can still use those individual pieces to get incremental results.  

WO: Can you tell us more about how we can choose to improve efficiencies and safety with this service?  

JS: First and foremost, safety is paramount. So, how can we move people out of harm's way? On the rigs, you have red zones, and you need to keep them out. For years, we've been doing remote monitoring. That means you have people looking from a remote center, anywhere in the world, overlooking from troubleshooting advisory services, but not really in control. You still have the same amount of people at the rig site.   

Now, when you move to autonomous drilling, we can actually start moving those people off the rig site itself. Now, you don't need as many people locally on-site making decisions, because decisions are already pre-programmed, and the algorithm has its own self-adjusting mechanism. From that standpoint, we're taking humans out of harm's way and putting them back, so they can still monitor jobs and work on the iterations of the algorithms.   

Fig. 3. Exemplified by the PermaFORCE-TRU drill bit, the Kantori™ service will improve bit performance while drilling, yielding a better-quality hole. Image: Baker Hughes.

I mentioned on the environmental side, that all these things do reduce rig time, which enables us to then reduce the carbon footprint, which is important for the industry. On the ROP, with the AI piece, it's constantly making micro-adjustments. So before, you would have had to continually adjust your profile, and think about gathering information downhole. Then, you're having to uplink it with mud pulse telemetry, and somebody will have to take it, look at it, and analyze it. Then, you have to stop uplinking, downlink, and control the tool. And the whole time that you're drilling, you're going off course.  

Well, now, if you're doing that at the bit (Fig. 3), you'll be able to iterate faster, and stay there, so you have a better-quality hole. You’re able to extend the reservoir contact section, and it's a better borehole quality, when you go back in with your completion or whatever comes next. It's a nice hole, to where you no longer have the situation where you've drilled it and now you can't complete it. Or you've drilled it, and now you can get the production off the entire interval. It maximizes your investment by making sure that hole quality is in the right spot. And with a quality that's good for the next steps.   

WO: How is AI enhancing operations and autonomous drilling, and how do you see it sensing the direction of travel, as well?   

JS: For our Kantori™ system, it's got really advanced algorithms, we have a physics-based model, and we have A.I. So, A.I. means a lot of things to a lot of people, but it's actually a culmination of a lot of different things coming together. A true benefit is making decisions quickly, in a non-subjective manner. When you think about how we train people for decades, where you hire somebody, the longer the tenure they had, the better decisions they could make through experience. We're now able to take decades of experience and boil it down, and we get more consistency of service. So that's what you get from the A.I. It's the ability to make micro-adjustments, real time, without having to have a 25-year drilling hand that's been in that application.   

The other thing, as this is still relatively young in the grand scheme of things, you really need to build up the data set to be able to continually integrate. One thing that makes it possible is these large data sets that are now harvesting more and more information. This allows A.I. to roll that down to not only what did I do well for this well, but now what can I do to scale it, to where if I learn something in Thailand, in short order, I could be applying it in Brazil. And you're not doing the telephone. Lastly, it provides that consistency of service to where we're all acting the same, which elevates the global bar further.   

WO: What would you say are the biggest challenges in autonomous drilling?  

JS: I think the first one is probably adoption. I link it to self-driving cars—do you really want to take your hand off the wheel? It sounds good until you want to do it. So, it’s from a standpoint of allowing the system to work, as designed, versus I'd like it to advise me, but I still want to make the decision. There's a hearts-and-minds cultural piece that's a real barrier for speed of implementation. When you take a look at our customer base, we have some that are very welcoming to that, and we have some that are very much wait-and-see. I'm not really ready to take that on.   

So, I think the first barrier is we have the technology, and we take a look at how it's built. It's really a culmination of a lot of things just married together and elevated with some computing horsepower. But again, everyone has a different risk tolerance for management of change, so that would be one. When you really talk about A.I. and all of what we're doing, you get better through repetition. So, the more experiences we have, the better the outcome. It's really about a volume game to build the data sets up to learn and to repeat.   

WO: So basically, this gives those customers a reassurance that they've built up the data set, and they have the experience that you are maybe concerned they don't quite have.   

JS: There's some of that, and there's also some contingency to where, if it doesn't work as designed, what can happen? We still have people in remote centers, where if something were to happen—loss of connectivity for instance—that you can still do the job or you're not going to have a blowout or anything else. It's more about talking about the benefits, but if something were to happen, here's the contingency, and it's going to be okay. But let's go ahead and go through it and learn together. So, that's one.   

The second one, I would say, is probably the rig interfaces. Obviously, we work with a lot of different rig providers. They all have their own software systems and control interfaces. Because they're different, it's not a one-size-fits-all. We have to configure as we go. We start with the largest rig suppliers, and it takes some time to do that. And you also have rigs without those interfaces, to where it doesn't exactly marry up. That's another big concern in how we move forward.   

WO: How does autonomous drilling fit into overall operations?  

JS: Obviously, on the front drilling side, everything has its place. Certainly, the more complex geometry is when you're building angle, when you do an extended reach, when you are in an environment where historically you maybe had to trip bottomhole assemblies out, just due to wear and tear. Now, we can get more longevity out of it. It helps reduce the cost of ownership, just because we don't have as much fleet replacement, and the annual costs go down. Again, it's a separate way of doing things with a higher efficiency output. So fewer people, fewer emissions, less wear and tear, better hole quality, better well placement, and ultimately, all that combines into reservoir contact and producing barrels of oil or gas equivalent at a lower price per unit. And at a low carbon set. That's the value proposition where it hits.   

WO: Is there a particular region or field type that’s more suited to this service, or does it work in a wide variety of situations?  

Fig. 4. The beauty of the Kantori™ service is that even if an operator is drilling shallow vertical wells, some of the pieces of the service will still harvest data from those wells to learn. Image: Baker Hughes.

JS: That's a good question. It's just like anything else—it really comes down to an economic and risk model. It is applicable in all geographies around the globe, but it is reservoir-dependent. In other words, how deep are you? What does the geometry in the survey of the well look like? Have you had problems in the bore before with stuck pipe, hole collapse, things like that? Because if you start sticking BHAs, and now you've not only jumped the well, but you have to buy the tool and we lose the tool, that's not good for anybody.   

So, how can you be predictive about risk mitigation and hazards in the well to prevent it? I would say again, thin layers, places with high cost of operatorship, complex completion design, complex well design. But certainly, if you're drilling a shallow vertical well, it's probably not the choice. Again, some of the pieces will still harvest data from those wells to learn, but the full Kantori™ (Fig. 4) probably wouldn't have the same benefit as if you had a long, deviated horizontal and complex geometry wellbore.  

WO: How do you see autonomous drilling evolving in the industry? 

JS: It's definitely a way of the future. It's not necessarily if, but when does it converge. The one thing you see when you look at the industry as a whole, there obviously is a lot of competition with more energy sources coming on. Hydrocarbons are always under pressure to become more cost-effective energy to the world. We've got the carbon piece in there, as well. I think one of the things we haven't done as well as an industry is the consistency of service. You have to know, does it work here, there, or wherever. We certainly have areas where we have great efficiency, great optimization, great productivity. We have other areas that we don't do as well on.  

What it really allows you to do is achieve that consistency of service, to where you can replicate a winning formula over and over and over again across the world. And then, as I mentioned, previously, you had to have a 20-to-25-year hand to really be an expert. Now, you're letting the machine learning do all the heavy lifting and repeatability. This allows us to scale with a less-experienced workforce than what we did in the past. This enables us to be more flexible, as we grow with energy demands.  

WO: Just to follow up and play devil's advocate a little bit. Is there a risk then, that the A.I.  might take learning and experience-building away from a lot of young people coming in? Where and how do people build their experience, or is it purely in the role of technology?  

JS: I think there's a role for people everywhere. I don't think we have an equation, where we're removing them completely out of it. There are still pieces, where we want to remove people out of harm's way, but you still take a look at role planning. You take a look at how we work with our customers, and if they have different goals. So, there are some interfaces that you're not going to replace with A.I.  

But you bring up a good point about the technology side. We have Kantori™ launching it, but it's certainly not the end result. It's a piece on the road to continual process improvement in continuous ways to make energy safer and more affordable. It's all about keeping our people ideating, thinking about it, working with our customers on meeting their goals and KPIs. But also taking them out of harm's way and taking them away from the redundant acts. It's the redundancy that we really want to turn over to digital, but the innovation, the thought leadership, that's what we want our intellectual core to start repeating around.  

JIM SESSIONS has been with Baker Hughes for 25 years and currently serves as Global Senior Vice President for Well Construction under the Oilfield Services & Equipment (OFSE) Business Segment. His management responsibilities span globally throughout multiple countries and continents, including the Americas, Europe, Middle East, Asia, and Africa. In his previous roles, Mr. Sessions has held leadership and management positions in Operations, Technology, Global Sales, Product Management, Integration and Supply Chain in North America, Russia, and Southeast Asia. Prior to his transition to lead the global well construction team, he was Senior Vice President for the Sales & Commercial teams within OFSE. Mr. Sessions graduated from Texas A&M University with a bachelor’s degree in mechanical engineering.