WillCo Drilling Intelligence - Edition 24: Autonomy Earns Its Authority

Figure 1. Autonomy is earned one rung at a time: WillCo's grouping of this week's ADIPEC papers and fleet news, from shadow mode to fleet infrastructure, with each paper rung carrying the criterion its paper reports and the top rung previewing the fleet stories below.

🛢️ Top Story

Four ADNOC Papers, One Qualification Ladder for Closed-Loop Drilling

Published November 2 in the ADIPEC 2026 proceedings as the conference opened (November 2 to 5).

Four ADNOC papers climb the ladder from advisory to autonomy: an accuracy gate before shadow mode, 80 to 98% autonomous footage in the curve, and fully autonomous geosteering enabled only after the closed loop proved consistently accurate.

Shadow mode to advisory (SPE-235314-MS): gate, a proof-of-concept (PoC) acceptance criterion of median absolute percentage error (MdAPE) below 15% before shadow mode; result, ADNOC Onshore's AI rate of penetration (ROP) advisor met all pilot success criteria on 5 wells and 16 hole sections drilled in advisory mode on 2 rigs.

The machine-learning (ML) ROP models, developed by ADNOC Onshore and three partners and trained on data from 70 offset wells, returned an MdAPE of 6.5% offline and 3.8% online in blind tests, then ran in shadow mode on live data in an in-country cloud before advisory deployment.

Higher adherence to recommended setpoints was generally associated with ROP gains; trajectory control and vibrations constrained the gains in build-dominated and dense intervals. The authors frame advisory as a structured qualification stage toward closed-loop drilling automation.

Advisory with crew trust (SPE-236142-MS): 8 days saved versus typical rig performance across three consecutive wells with different designs, with more than 75% adherence, ADNOC Onshore and SLB report. Parameter execution became more consistent between shifts as operational trust increased throughout the campaign.

The pilot rested on three pillars: crew onboarding reinforced by daily reviews, the Real-Time Operations Center (RTOC), and drilling subject matter experts (SMEs). The authors describe the results as record-setting, including the fastest ROP in the 12.25-in., 8.5-in. and 6-in. sections, with reduced invisible lost time (ILT).

Autonomous directional, curve section (SPE-235625-MS): 80 to 98% autonomous footage; total downlinks down 12 to 46% and off-bottom downlinks down 83 to 93%; 42 hours of combined rig time saved across two of the three wells. ADNOC Onshore and SLB call it the first successful pilot of autonomous directional drilling (ADD) in a UAE onshore field, in the curve section on complex 3D trajectories.

Closed-loop geosteering, ADNOC Offshore (SPE-235299-MS): gate, several monitored iterations until the system was consistently accurate; result, fully autonomous geosteering enabled, ADNOC Offshore and SLB authors report, with analysis still ongoing. The chain runs from real-time logging-while-drilling (LWD) data and an ultra-deep resistivity (UDR) inversion in the cloud to the rotary steerable system (RSS), in what the authors call a fully closed loop in a conductive reservoir:

LWD data → cloud UDR inversion maps reservoir boundaries → ML picks the target → forward plan ahead of the bit → ADD downlinks to the RSS.

🎯 Your move: before your next advisory pilot, write down two numbers: the one that promotes the system to the next rung and the one that sends it back (WillCo analysis).

📩 Not sure which of your gates are written down? Check them against the free 15-question Closed-Loop Readiness scorecard: willcodrilling.com/closed-loop-readiness.

🎛️ Who Holds the Veto

ADNOC Onshore and NOV Close the Loop on a Chinese Rig Control System Without Moving the Interlocks

The optimizer sends weight on bit (WOB) and rotary speed (RPM) setpoints; the rig control system (RCS) keeps command validation, operating limits and safety interlocks; the driller can resume manual control immediately whenever required.

ADNOC Onshore and NOV Digital Services report what the authors describe as the first documented integration of an AI drilling optimizer with a Chinese rig control system, running fully closed-loop optimization of drilling parameters (SPE-236489-MS). A layered communication framework separates the optimization algorithms from the control system interfaces, so neither the optimizer nor the drilling control logic had to be fundamentally modified. The engine reads real-time measurements from the rig control network and returns its optimized setpoints through a secure interface.

Figure 2. Who holds what in the ADNOC Onshore and NOV integration: the optimizer proposes, the rig control system keeps command validation, limits and interlocks, and the driller can override.

The team applied a structured engineering methodology that runs through hardware-in-the-loop (HIL) validation and factory acceptance testing (FAT) before the rig:

1 Communication mapping → 2 interface verification → 3 simulator testing → 4 HIL → 5 FAT → 6 site commissioning → 7 phased operational deployment.

The engineering emphasis fell on deterministic communications, clear transfer of control ownership between the automation and the driller, fail-safe behavior on communication loss, and verification that automated commands could not exceed predefined limits. Field deployment showed stable closed-loop operation. The authors conclude that differences in controller architecture, software organization and human-machine interface (HMI) design can be accommodated through standardized communication interfaces, and they call the project a milestone toward vendor-independent autonomous drilling.

Offshore, on jack-up Rig J (SPE-236491-MS): ADNOC Offshore upgraded jack-up Rig J's RCS, plus additional software, in four phases: offline preparation, commissioning, shadow mode, then the autonomous scope, covering on-bottom and slip-to-slip operations.

🎯 Your move: ask your contractor to show in writing which system owns each limit and interlock once an optimizer is connected, and what the rig does in the first seconds after communication is lost (WillCo analysis).

🔌 Fleet Plumbing

Halliburton Buys Keystone, NOV Wins a Fleet-Wide Max Award, Nabors Extends RZR, and an ADNOC Drilling Follow-up

Halliburton acquires Keystone and its SmartHUB integration layer, NOV reports a supermajor fleet-wide Max award, Canrig commitments will take RZR to six deployed modules, and, in a follow-up, ADNOC Drilling's nine-month results list AD-300 and AD-301 as deployed.

Halliburton + Keystone (November 2): Halliburton has acquired Norway-based Keystone AS, whose SmartHUB integration layer, well-planning workflow and rig-execution tools, Halliburton says, extend Landmark's AI-native well-delivery ecosystem from the office to the rig; Halliburton disclosed no deal value.

NOV Max, supermajor fleet award (reported October 27, third-quarter results): NOV reports a Max Platform award for data aggregation and remote monitoring across a supermajor's global offshore fleet of drillships, jackups and drilling platforms, which NOV says will lay a scalable digital foundation for future automation, alongside a drillship upgrade award for NOVOS and Multi-Machine Control.

Canrig RZR, to six modules (reported October 27, third-quarter results): The same day, Nabors said Canrig secured commitments for four more RZR rig floor automation modules, which will bring the total deployed to six.

Follow-up, ADNOC Drilling AD-300/AD-301: ADNOC Drilling's nine-month results, also October 27, list the deployment of AD-300 and AD-301, its first AI-enabled automated island rigs, alongside four additional planned rigs.

🎯 Your move: before signing any fleet data contract, confirm the data model and its event definitions are yours to export if you change vendors (WillCo analysis).

🧠 Governing the Agents

Pions and Eni Ground Drilling Agents in a Governed Layer; SLB Benchmarks a Time-Aware Memory

SLB's agent got every as-of question right once it had a time-aware fact table; retrieval over a hand-designed schema returned confident false negatives.

Pions + Eni (SPE-236023-MS): Pions and Eni start from a blunt premise: as operators deploy drilling agents, the binding constraint is no longer model capability but context, and an unconstrained agent will produce confident, physically impossible recommendations. Their answer is a knowledge graph over a governed drilling data layer. Every answer carries an explicit evidence chain. Governed queries return in seconds and complete agent responses in one to two minutes, the authors report.

The guardrails sit in a harness around the agents rather than in the model: agents reach the layer only through governed tool interfaces, where the harness enforces physics-range checks and an evidence and confidence hierarchy, and it puts human-in-the-loop gates on hard-to-reverse actions. The ontology is aligned to OSDU and Energistics WITSML; on a five-well dataset, the pipeline materialized 806 nonproductive time (NPT) events with zero load failures.

SLB (SPE-235330-MS): with a bi-temporal fact table and a point-in-time query tool, SLB's agent reached 100% as-of accuracy on both corpora, answering every "what did we know on that date?" question correctly; all nine static errors were corrected, with no regressions across 96 paired queries. SLB built and tested it on two public corpora: 1,681 daily drilling reports (DDRs) from Equinor's Volve field and 1,888 Bureau of Safety and Environmental Enforcement (BSEE) weekly activity reports from Shell Offshore in the Gulf of Mexico.

A schema audit found that a hand-designed schema hid named storms that appeared in 30 reports, producing what the authors call confident false negatives.

🎯 Your move: make every agent answer cite the records it used, and reject any answer that cannot (WillCo analysis).

📏 The Measurement Under the Loop

Sinopec Tech Middle East and Tomax Show What a Loop Can and Cannot See

A single density rule ran 139 seconds late on a gradual mud event (SPE-235826-MS), and Tomax notes that surface loads are typically recorded every 10 seconds or longer, while its pipeline runs at 4,000 Hz.

Sinopec Tech Middle East benchmarked four anomaly detectors on a 48-hour online drilling fluid monitoring field record: a Shewhart and cumulative sum (CUSUM) rule on density, a multivariate change-point test, an isolation forest and a one-class support vector machine (SVM). All four recovered all three density-collapse events; the gradual event's onset was led by a decline in plastic viscosity. The density rule caught the sharp events at onset, while the multivariate detectors flagged the gradual one at or before onset.

Nils Reimers of Tomax and Ole Andre Høiland, an adviser to Tomax, set out the surface-rate contrast in a JPT case study on November 1. Their data come from a strain and torsion sub on the derrick drilling machine of an automated drilling rig built in late 2025. On a geothermal appraisal well in gneiss, a four-hour, three-BHA rainflow comparison amounted to 115 million calibrated datapoints per trace.

Figure 3. A loop can only act on what its measurement sees, in resolution and in time.

🎯 Your move: on your next automated well, replay one week of mud alarms against a multivariate confirmation rule, and ask what sample rate your vibration limits were set on (WillCo analysis).

📋 In Brief

• 📈 31% faster lateral ROP from automated surface pressure management - E3 Company (SPE-235414-MS): on a five-well pad, against offsets drilled on the same rig. During an internal blowout preventer (IBOP) malfunction, the system detected two consecutive overpressure events and limited peak surface pressure.

• 🧪 IPT Global takes blowout preventer (BOP) testing digital on a jack-up (SPE-235628-MS): more than 150 pressure tests in seven months for a major operator; critical-path BOP test duration fell nearly 50%. Some tests passed decline-rate thresholds but failed thermodynamic analysis. 🎯 Your move: which of your rig's tests would pass on decline rate alone? (WillCo analysis)

• 📍 A wireline run worth 12 to 24 hours of NPT replaced by real-time LWD collar location - Halliburton and Eni US (SPE-235871-MS): across three wells, LWD replaces the wireline casing collar location (CCL) run; first-pass detection succeeded in all three, and the planned second pass was cancelled. The authors call it a world-first solution, applied in the Gulf of America. 🎯 Your move: which wireline runs in your next program exist only to confirm a depth? (WillCo analysis)

• 🔍 ADNOC Offshore and SLB image the formation at the bit (SPE-235469-MS): in what the authors describe as the first offshore field application of at-bit formation imaging, in the UAE, two stinger elements behind the polycrystalline diamond compact (PDC) cutters on a 12.25-in. bit resolved bedding planes and vugs in a Cretaceous carbonate, independent of mud type.

• 🚨 ADNOC Offshore and SLB catch drill string washouts early (SPE-236138-MS): a physical model with Bayesian linear regression, applied on all wells drilled, detected all the early washouts, the authors report; in a 6-in. RSS BHA example, alarms led to a trip out that found washouts at the LWD tool.

• 🧭 SLB scores section risk with an agentic large language model (LLM) (SPE-235786-MS): a trajectory similarity index and SME weights produce a 0 to 100 risk score for each section, cutting offset review from days to minutes; the method surfaced hidden incidents not labeled as NPT, up to 50% relative to labeled NPT alone in some Volve iterations, the authors report.

• 🧫 Mud data delay cut from 24 hours to under 2 minutes across 11 rigs in Ghawar (2025 paper SPE-227670-MS, highlighted by JPT on November 1): automated rheology and density testing in Saudi Arabia's Ghawar gas field, across 48 wells and more than 300 sections, streamed over WITSML; the authors report more than 45 documented remote interventions and seven well control events mitigated. 🎯 Your move: how old is the last mud check when the driller acts on it? (WillCo analysis)

• 🏗️ Patterson-UTI on what customers now specify (October 28): CEO Andy Hendricks said customer programs are increasingly requiring upgraded rigs with advanced digital and automation capabilities and larger structures designed to drill deeper wells and longer laterals more efficiently. 🎯 Your move: check whether your next rig tender specifies automation capability or only rig class (WillCo analysis).

📡 On WillCo's Radar

Why it matters: the oversight clause is the line to watch: which planning decisions count as consequential, and who signs them, will decide how far the speed claim travels toward the rig (WillCo analysis).

• data² launches evidence-graded well performance at ADIPEC (November 2): data² says its reView platform gives every value one of six trust tiers, from measured and verified down to inferred, and that an unchecked value can never drive a ranking. Business records follow PPDM 3.9, with new data arriving in WITSML and PRODML with its chain of custody attached. Read: trust tiers are the evidence chain applied to well data (WillCo analysis).

• dss+ on human-factor risk at a distance (SPE-235732-MS): Graham Sheedy argues that safety frameworks are not yet fully adapted to remote and autonomous operations and proposes a structured suite of controls. Read: the safety case for remote and autonomous operations is still being written (WillCo analysis).

• DeKUT on what surface data cannot see (published October 27): a simulation study across 90 operating points finds wearflat bit friction recoverable from surface torque, top drive speed and hook load at 58% of stick-slip points but only 4% of smooth ones. Read: smooth drilling is where surface data says least (WillCo analysis).

• Stanford reconstructs logs from cuttings and drilling data (SPE-236466-MS): the paper's feature-based multi-well model, trained on 14 Utsira High wells, reached a compressional sonic (DT) R² of 0.93 on a blind well; the multimodal fusion model scored 0.858 on three blind wells. Read: cuttings plus drilling data become a log source where wireline is skipped (WillCo analysis).

• Corvacon follow-up (carryover: event October 20, recap October 29): the October 29 recap, a follow-up to Edition 22, adds the "Connected by Corva. Owned by You." ownership stance; it attaches no measured field result yet. Read: "Owned by You" is a data-ownership pitch, the same contest as Fleet Plumbing (WillCo analysis).

• Patterson-UTI's Digital Performance Center (Drilling Contractor, published October 27): Part 1 of a three-part video series presents the Houston center as the real-time link between field and office. Read: drilling contractors, not only operators, now run their own field-to-office link (WillCo analysis).

🔭 What We're Watching Next

• ADIPEC, through November 5: further autonomy papers and announcements from the technical programme.

• Patterson-UTI: Parts 2 and 3 of the Drilling Contractor Digital Series; its third-quarter release and call gave no fleet automation metric to test the auto driller figures in the SPE-234119-MS paper covered in Edition 23.

• ADNOC's ladder: results from the sections drilled after the Rig J upgrade, which the abstract says are presented but does not quantify, and the outcome of ADNOC Offshore's ongoing autonomous geosteering analysis.

What has to become standardized before automation can become routine? For the rig-control layer, the ADNOC Onshore and NOV authors already answer: standardized communication interfaces.

The next commercial advantage is not a better optimizer but the qualification package that lets any optimizer be trusted on any rig: written gates, shared event definitions and a veto that stays with the safety case.

Read across the week, three things need the same treatment (WillCo analysis). The gate: ADNOC's teams moved from an accuracy criterion to shadow mode, then advisory, then autonomy only after the system proved consistent, and none of the abstracts points to a shared industry convention for those gates, which leaves each operator writing its own. The data: Keystone's integration layer, NOV's fleet-wide Max award, data²'s trust tiers and the Pions governed layer are all attempts to make a rig event mean one thing everywhere. The veto: the rig control system kept the interlocks, the driller kept the override, and the ADNOC release says human experts will retain oversight of consequential decisions.

Figure 4. The gate, the data and the veto: the three control questions this edition's stories leave operators, applied by layer from planning to fleet scale (WillCo analysis).

The tension is pace. At the planning layer ADNOC says it will deliver master plans in hours rather than months; at the rig layer the same company qualifies a drilling loop through shadow mode, pilots and phased deployment. Both can be right: speed is earned where a wrong answer is reversible, and the ladder governs where it is not. The measurement papers set the floor under all of it, because a loop acting on slow or single-variable data inherits that data's blind spots however well its gate is written (WillCo analysis).

✅ Practical Recommendations

For operators and drilling contractors (WillCo analysis): treat autonomy as a qualification program, not a product purchase. Start with Recommendations 1 and 2: neither needs new technology.

1. 🎛️ Map the veto this week: one page per rig showing which system owns limits, interlocks and command validation, how control ownership transfers, and how the driller returns to manual.

2. 🪜 Sign the gate into the pilot contract before mobilization: have the accountable drilling and contract owners sign the acceptance criterion for each rung (accuracy before shadow mode, a shadow-mode run before advisory, a benchmark to judge advisory, demonstrated consistency before autonomy) into the contract before the rig mobilizes.

3. 🔌 Define your core events fleet-wide before the next platform renewal: agree on the definitions of connections, trips and alarms before buying another platform, and write them into every data contract.

4. 🧠 Test agents on time in the next agent pilot: include as-of questions in every agent acceptance test, and require an evidence chain behind every answer.

5. 📏 Audit channel rates this quarter: list the sampling rate and latency of every channel a loop acts on, and flag single-variable alarms watching for gradual events.

A loop that earns its authority in writing gets to keep it.

📩 Start this week: map the veto on one rig. It takes one page and no new technology. Then score that rig on the free 15-question Closed-Loop Readiness scorecard.

📚 References and Sources

1. S. G. Rachman, K. Alissa, M. S. Ashraf, M. AlNuaimi, B. N. Kodiah, A. Galimkhanov, P. Sabhapandit, A. Khoshnaw, K. Singh, G. Marwah, F. S. Boukdera, W. Pavsin, N. Porwal, S. Balakrishnan. "Scaling AI-Enabled Drilling Optimization Solution for Drilling Activity in the United Arab Emirates: Advisory Workflow Field Validation Toward Closed-Loop Drilling Automation," SPE-235314-MS, ADIPEC 2026. DOI: 10.2118/235314-MS. https://doi.org/10.2118/235314-MS

2. M. Jalbout, A. Al Hai, I. Ahmed, K. Alissa, A. Elbagoury, M. Saad (ADNOC Onshore); A. Babakov, A. Qadir, G. Hernandez, A. Siddique, H. Nuri (SLB). "Driving Record-Breaking Drilling Performance Through the Implementation of a Drilling Advisory Solution: A Three-Well Pilot Case Study," SPE-236142-MS, ADIPEC 2026. DOI: 10.2118/236142-MS. https://doi.org/10.2118/236142-MS

3. B. Hernovianto, M. A. Al Nuaimi (ADNOC Onshore); A. E. Amin (SLB). "Autonomous Directional Drilling Pilot Implementation in the United Arab Emirates Onshore Field," SPE-235625-MS, ADIPEC 2026. DOI: 10.2118/235625-MS. https://doi.org/10.2118/235625-MS

4. O. Mohammed, S. Yousif, K. A. Alhosany, S. A. Ahmed, P. R. Agrawal, T. Saqib (ADNOC Offshore); K. El-Derini, S. Osman, W. Siddiqui, J. Kattukuzhiyil, S. Basso, R. Gomes, R. Aubakirov, D. Bourgeois (SLB). "Autonomous Geosteering: Integration of Automated Subsurface Interpretation and Evaluation, Direction Well Planning and Drilling in a Fully Closed Loop System," SPE-235299-MS, ADIPEC 2026. DOI: 10.2118/235299-MS. https://doi.org/10.2118/235299-MS

5. A. O. Akinlosotu, R. A. Al-Rahma, M. A. Taha, I. M. Ahmed, R. H. Alhammadi, M. A. Al-Mohammad, M. Reyami, A. I. Eissa, A. M. Ibrahim (ADNOC Onshore); R. Serrano, R. Abboud (NOV Digital Services). "First Global Deployment of a Closed-Loop AI Drilling Optimizer Integrated with Rig Control Systems," SPE-236489-MS, ADIPEC 2026. DOI: 10.2118/236489-MS. https://doi.org/10.2118/236489-MS

6. D. Yudhia, H. Aljaberi, Y. Zhao, R. Maarouf, P. Bimastianto, S. Khambete, M. Afzal, M. Al Ali, T. Almarzooqi, A. Alharbi, M. Abdelsattar, M. Khadim, S. Alameri, S. Al Hajri, S. Alblooshi, Y. Alhammadi, Y. Al Khateeri (ADNOC Offshore). "The Journey to Achieve Autonomous Drilling in Jack Up Rig Operations of Offshore Abu Dhabi," SPE-236491-MS, ADIPEC 2026. DOI: 10.2118/236491-MS. https://doi.org/10.2118/236491-MS

7. "Halliburton acquires Keystone AS to strengthen digital well construction capabilities" - Halliburton press release, Nov 2, 2026. https://www.halliburton.com/en/about-us/press-release/halliburton-acquires-keystone-strengthen-digital-well-construction-capabilities

8. NOV third-quarter 2026 results, Exhibit 99.1 to Form 8-K, Oct 27, 2026. https://www.sec.gov/Archives/edgar/data/1021860/000119312526435237/nov-ex99_1.htm

9. Nabors third-quarter 2026 results, Exhibit 99.1 to Form 8-K, Oct 27, 2026. https://www.sec.gov/Archives/edgar/data/1163739/000110465926119167/tm2628332d1_ex99-1.htm

10. ADNOC Drilling third-quarter and nine-month 2026 results, Oct 27, 2026. https://adnocdrilling.ae/en/news-and-media/news-releases/2026/3q-2026-results

11. J. Cao, A. Livincovs, T. Yu, R. B. Muhammad (Pions); L. Bianchi, L. P. Bianchini, A. Ciuca, V. Gibilterra, G. Sinibaldi (Eni SpA). "Grounding Intelligent Drilling Agents: A Knowledge Graph Over a Governed Drilling Data Layer for Verifiable Agentic Workflows," SPE-236023-MS, ADIPEC 2026. DOI: 10.2118/236023-MS. https://doi.org/10.2118/236023-MS

12. A. Bihani, S. Manikani, H. Arabi, M. Liu (SLB). "Knowledge Organization Matters: How Graph Design and Temporal Structure Shape Agentic AI Performance on Drilling Activity Reports," SPE-235330-MS, ADIPEC 2026. DOI: 10.2118/235330-MS. https://doi.org/10.2118/235330-MS

13. Rached, Rodriguez-Hernandez, Davalos-Monteiro, Pensa, Zuqing (Sinopec Tech Middle East). "When is a Density Rule Enough? a Field Benchmark of Anomaly Detectors for Real-Time Mud Monitoring," SPE-235826-MS, ADIPEC 2026. DOI: 10.2118/235826-MS. https://doi.org/10.2118/235826-MS

14. N. Reimers (Tomax); O. A. Høiland (AD Navigation, a Jotron subsidiary; adviser to Tomax). "Case Study: AI Redefines Next-Gen Drilling Vibration Compliance" - JPT, Nov 1, 2026. https://jpt.spe.org/case-study-ai-redefines-next-gen-drilling-vibration-compliance

15. J. Clark, A. Epead, A. El Beltagy, C. Fuser, J. Archuleta, B. Shuman (E3 Company). "31% Lateral ROP Uplift Through Remotely Automated Surface Pressure Management: Field Results from U.S. Unconventional Drilling," SPE-235414-MS, ADIPEC 2026. DOI: 10.2118/235414-MS. https://doi.org/10.2118/235414-MS

16. C. Rennie, C. MacDonald, J. Sutler (IPT Global). "Lessons from the Collaborative Adoption of Digital Pressure Testing Technology to Improve Offshore Drilling Efficiency and Barrier Assurance," SPE-235628-MS, ADIPEC 2026. DOI: 10.2118/235628-MS. https://doi.org/10.2118/235628-MS

17. R. Berama (Halliburton); N. Sharma (Eni US Operating). "A World-First Real-Time LWD Casing Collar Location (CCL) Solution for Dual-Casing, Vertical and High-Angle Sidetrack Operations in the Gulf of America," SPE-235871-MS, ADIPEC 2026. DOI: 10.2118/235871-MS. https://doi.org/10.2118/235871-MS

18. A. Serry et al. (ADNOC Offshore, SLB). "First Offshore Deployment of At-Bit Formation Imaging Globally in the Middle East: Field Introduction and Learnings from the UAE," SPE-235469-MS, ADIPEC 2026. DOI: 10.2118/235469-MS. https://doi.org/10.2118/235469-MS

19. S. Duivala et al. (ADNOC Offshore, SLB). "Early Detection of Drill String Washout Using Real-Time Abnormal Pressure Monitoring and Automated Alarms," SPE-236138-MS, ADIPEC 2026. DOI: 10.2118/236138-MS. https://doi.org/10.2118/236138-MS

20. P. Chica, C. J. Sena Santiago (SLB). "Trajectory-Aware Drilling Risk Scoring: An Agentic LLM Approach to Offset Well Selection and Dynamic Section Risk Assessment," SPE-235786-MS, ADIPEC 2026. DOI: 10.2118/235786-MS. https://doi.org/10.2118/235786-MS

21. Dorantes, Ghali, Torres, Atencio, Moreno, Corona. "Enhancing Drilling Efficiency: Real-Time Rheology and Density Profiling with RheoProfiler in Saudi Arabia's Ghawar Gas Field," SPE-227670-MS, SPE Middle East Oil and Gas Show 2025. DOI: 10.2118/227670-MS. https://doi.org/10.2118/227670-MS ; JPT highlight, Nov 1, 2026: https://jpt.spe.org/automated-rheometer-enhances-efficiency-in-saudi-arabias-ghawar-gas-field-restricted

22. Patterson-UTI third-quarter 2026 results, Exhibit 99.1 to Form 8-K, Oct 28, 2026. https://www.sec.gov/Archives/edgar/data/889900/000088990026000070/a2026q3earningsreleaseex991.htm

23. "ADNOC Signs Landmark Partnership Agreement with OpenAI and G42, Becoming First Energy Company to Deploy OpenAI Frontier Technology" - ADNOC press release, Nov 2, 2026. https://www.adnoc.ae/media-center/news/2026/11/ADNOC-Signs-Landmark-Partnership-Agreement-with-OpenAI

24. "data2 Launches Asset Performance & Optimization for Energy: Well Performance Backed by Evidence You Can Trust" - data² via Presswire, Nov 2, 2026. https://presswire.com/release/data2-launches-asset-performance-optimization-for-energy-well-performance-backed-by-evidence-you-can-trust/

25. G. Sheedy (dss+). "Operating at a Distance: A Framework for Managing Human-Factor Risk in Remote and Autonomous Energy Operations," SPE-235732-MS, ADIPEC 2026. DOI: 10.2118/235732-MS. https://doi.org/10.2118/235732-MS

26. J. Petu, C. Kagiri, S. M. Kang'iri (Dedan Kimathi University of Technology). "Observability Limits of Downhole State Estimation from Surface Measurements in Geothermal Drilling," Book of Abstracts, 10th DeKUT International Conference on Science, Technology, Innovation and Entrepreneurship (STI&E 2026), Oct 27, 2026. https://doi.org/10.33803/dgzvrs

27. A. Elgamal, T. Mukerji (Stanford University). "Multimodal AI Reconstruction of Petrophysical-Scale Logs from Drill Cuttings and Drilling Data: A Sonic- And Density-Validated Framework Toward Quad-Combo Formation Evaluation," SPE-236466-MS, ADIPEC 2026. DOI: 10.2118/236466-MS. https://doi.org/10.2118/236466-MS

28. "Corvacon: Nexus 2026 Recap" - Corva blog, Oct 29, 2026. https://www.corva.ai/blog/corvacon-nexus

29. S. Whitfield. "Patterson-UTI Digital Series, Part 1: Performance center serves as real-time link between field and office" - Drilling Contractor, Oct 27, 2026. https://drillingcontractor.org/patterson-uti-digital-series-part-1-performance-center-serves-as-real-time-link-between-field-and-office-80380

30. Prior WillCo Drilling Intelligence editions and archive - https://willcodrilling.com/blog

About this report: The WillCo Drilling Intelligence Report is a weekly synthesis of verified developments in digital drilling and well construction, published by WillCo Drilling Consulting. Analysis, forecasts, and recommendations reflect WillCo's independent engineering perspective.

📩 info@willcodrilling.com | All editions

📖 Glossary of Acronyms