Kansas City Artificial Intelligence: 20 Opportunities for 2026

Where agentic systems, causal inference and machine learning will create the most economic value in Kansas City, ranked by a founder who has worked in the city’s AI community since 2016.

By Brian James Curry, Founder of Vector1 Research, Kansas City

In my last piece, The State of AI in Kansas City in 2026, I argued that the region has moved from chat wrappers to decision systems. The obvious next question is where. Which decisions, in which Kansas City industries, are worth handing to agentic systems, measuring with causal inference, or improving with machine learning?

This is my answer: twenty specific opportunities for Kansas City artificial intelligence, grouped by industry and scored by economic impact.

I’m Brian Curry. In 2016 I started the Kansas City Machine Learning Group, which became KC AI Lab and grew to more than 1,500 practitioners. I have spent more than twenty years leading analytics, data science and machine learning inside enterprises, and I now run Vector1 Research, an independent applied AI lab in Kansas City. Many of the opportunities below sit close to the work I have done inside enterprises, and some of them are why I wrote the open-source tools I mention along the way.

The economic case is simple. In the KC Rising metrics the Mid-America Regional Council tracks, Kansas City produced about $99,000 of GDP per job in 2022, against a benchmark average of about $107,000 for peer metros, and its growth in that measure from 2017 to 2022 was about half the benchmark pace (MARC). Output per job is a productivity measure, and better decisions are one of the most direct ways to raise it. Every opportunity on this list is a way to close part of that gap.

How I Scored the Opportunities

I scored each opportunity by the annual value of the decisions it improves, discounted by how ready Kansas City companies are to act on it. The twenty are grouped by industry, and the ones with the largest economic impact are pulled together after the list.

The working formula is V = N × v × Δ × r. N is the number of decisions a year, v is the value at stake in each decision, Δ is the realistic improvement a better method delivers, and r is readiness from 0 to 1: clean data, a system of record the model can act through, and a team that can govern it.

Take a regional shipper that tenders 200,000 truckloads a year at an average of $2,000 per load. If an agentic procurement system improves carrier selection and pricing by 2%, and the company’s readiness is 0.7, the expected annual value is 200,000 × $2,000 × 0.02 × 0.7, or $5.6 million. The numbers are illustrative, but the structure is the point: high-volume decisions with real money attached and a clean path to execution rise to the top.

Each entry below names the decision, the method that fits it (agentic systems, causal inference, machine learning or a mix), and the open-source tool from my own work that applies where there is one.

Freight, Logistics and Trade

Kansas City’s largest AI opportunity sits in freight, because no other local industry makes as many repeatable, money-bearing decisions. KC SmartPort, the logistics initiative of the Kansas City Area Development Council, describes the metro as the largest rail hub in the U.S. by tonnage, with four Class I rail lines, four intermodal parks, and two-day reach to 90% of the contiguous U.S. (KC SmartPort).

1. Agentic freight procurement and carrier selection

Shippers and brokers tender thousands of loads a week, and most carrier choices still depend on a planner’s memory and a rate sheet. An agentic system can price each lane against current market data, select carriers inside contract and safety rules, tender the load in the transportation management system, and escalate only the exceptions. This is human-on-the-loop work in its purest form, and it is the worked example above for a reason.

2. Intermodal yard, rail ramp and dock scheduling

The ramps and warehouse clusters around Logistics Park Kansas City in Edgerton and the metro’s rail yards run on tight sequencing: which container moves first, which door a trailer takes, which crew covers which shift. Optimization models paired with scheduling agents cut dwell time, and dwell time is money for every party in the chain.

3. Network design and inventory placement

The two-day reach that makes Kansas City attractive only pays if inventory sits in the right building. Machine learning demand forecasts feeding a network optimization model decide where each product should live and how much safety stock it needs. I wrote a practical guide to the geospatial side of this problem in The Engineer’s Field Guide to Geospatial Analytics, Route Optimization, and GIS.

4. Forecasting supply shocks in trade inputs

Export controls, tariffs and fragmented supply chains now move input prices faster than historical models expect. Forecasts that model those shocks directly, instead of treating them as noise, give procurement teams time to hedge or re-source. My open-source project tincast does this for tin and other AI-critical commodities with a daily fragmentation index, and the same approach applies to any input a Kansas City manufacturer or distributor buys on a global market.

Agriculture, Animal Health and Food

The region’s agricultural and animal health businesses face causal questions more than prediction problems: what a price change, a treatment or a promotion actually does. The KC Animal Health Corridor stretches from Manhattan, Kansas, to Columbia and St. Joseph, Missouri. It counts more than 300 companies and about 20,000 animal health jobs, and its companies account for 56% of worldwide animal health, diagnostics and pet food sales (KC ADC).

5. Grain marketing and pricing for co-ops and elevators

Hard red winter wheat still trades on a contract that carries Kansas City’s name: CME Group’s KC HRW wheat futures, which price the national supply of the crop (CME Group). The region’s cooperatives and elevators set basis bids off that market every day. Machine learning forecasts of local basis, combined with causal estimates of how bids change farmer delivery behavior, turn a judgment call into a measured pricing policy. I grew up in a small Kansas farming town, where this is a decision farm communities face every harvest.

6. Research and regulatory knowledge for animal health

Animal health companies carry decades of study reports, regulatory submissions and field data in documents no one can search well. Knowledge graphs built from that corpus let scientists and regulatory teams find prior evidence in minutes, and agents can draft the first version of routine submissions for expert review. My open-source framework MeaningFlow builds this kind of semantic model from embeddings, clustering and graph methods.

7. Marketing measurement for pet food and consumer animal health

Consumer animal health and pet food brands spend heavily across retail, digital and veterinary channels, and attribution dashboards routinely over-credit whatever ran last. Bayesian marketing mix modeling estimates what each channel actually contributed, with honest uncertainty. That is the core use case of Papilon, my open-source framework for causal modeling, Bayesian marketing mix modeling and simulation.

8. Biomanufacturing process optimization

In July 2026 the U.S. Economic Development Administration awarded $33.5 million to the Kansas City BioSecure Manufacturing Tech Hub, led by BioNexus KC, to grow biologics and biomanufacturing in the region (Rep. Sharice Davids). Biomanufacturing yields depend on many interacting process settings, which makes it a strong fit for designed experiments, causal models of process variables, and machine learning quality prediction. A new hub gets to build that data foundation correctly from the first batch.

Manufacturing, Engineering and Energy

Kansas City exports engineering to the world, and that gives AI applied here a reach well beyond the metro. The Kansas City Area Development Council reports that Kansas City firms design 20% of the world’s power systems and 13% of global telecommunications systems, and that nearly half of global sports architecture originates here (KC ADC).

9. Automotive production scheduling and supplier risk

The metro’s two major assembly plants show why supplier risk is a modeling problem. Ford’s Claycomo plant builds the F-150 and the Transit van. In October 2025, a fire at a Novelis aluminum plant in New York cut aluminum supply and forced F-150 layoffs at Claycomo, while the Transit line, which does not use aluminum bodies, was expected to keep running (KCTV5). GM’s Fairfax plant, which builds the electric Chevrolet Bolt, is scheduled to add a gas-powered Equinox in mid-2027 and a Buick compact SUV in 2028 (KCTV5). Machine learning models of supplier delivery risk, combined with scheduling agents that replan when a part is late, protect line uptime when it matters most.

10. Agentic design review for architecture and engineering firms

Black & Veatch, Burns & McDonnell, HNTB, Populous and other Kansas City design firms produce enormous volumes of specifications, drawings and code-compliance checks. Agents grounded in a firm’s own standards and past projects can check a spec against applicable codes, flag conflicts between disciplines, and assemble first drafts of routine documents for engineers to review. Because these firms design infrastructure worldwide, a productivity gain here is a Kansas City export.

11. Grid and data center load forecasting

Google’s Project Mica in the Northland is planned to draw up to 500 megawatts at full capacity, and it falls under Evergy’s large load tariff (Axios Kansas City). Loads of that size change how the regional grid has to plan. Probabilistic load forecasting, with causal models of how weather, prices and new large customers interact, is one of the highest-stakes machine learning problems in the region.

12. Predictive maintenance and visual inspection

Across the region’s plants, from food processing to metal fabrication, unplanned downtime and quality escapes are the costs that machine learning reduces most reliably. Sensor-based failure prediction and computer vision inspection are mature methods now. The opportunity for mid-sized Kansas City manufacturers is adopting them without building a data science department, which is a readiness problem more than a technology one.

Health, Finance and Telecom

Kansas City has deep technical workforces in health IT, financial services and telecom, and each one sits on years of transaction data that has rarely been used for causal decisions.

13. Hospital capacity, staffing and patient flow

The region’s health systems manage beds, operating rooms and nurse staffing against demand that swings by season and by day. Machine learning forecasts of admissions and discharges, paired with scheduling agents that propose staffing changes for a charge nurse to approve, reduce both overtime and diversions. Kansas City also has an unusually large health IT talent pool, built over decades around Cerner, now Oracle Health, which shortens the path from model to production.

14. Credit, fraud and pricing decisions in financial services

Kansas City is home to the Federal Reserve Bank of Kansas City, regional banks, and consumer finance and tax companies such as H&R Block. Fraud detection is already machine learning territory. The bigger untapped opportunity is causal: testing credit line policies, fee changes and offers with proper comparison groups, so pricing decisions rest on measured effects rather than on historical correlations shaped by who got the offer.

15. Claims and service triage in insurance

Insurers in the region handle large volumes of claims and service requests that follow repeatable patterns. Agents can gather documents, check coverage, route each case, and settle simple claims within set limits, while adjusters focus on disputed and high-value files. Every action needs an audit trail, which is where my research on Memory-Node Encapsulation, a structured episodic memory for agent systems, applies directly.

16. Network operations and churn in telecom

T-Mobile employs about 3,500 people on the former Sprint campus in Overland Park (AOL), and the KC ADC reports that Kansas City companies design 13% of global telecommunications systems (KC ADC). Network anomaly detection and automated remediation agents are natural fits. So is causal churn modeling, which separates customers who would leave anyway from customers a retention offer actually saves, so the company stops paying to keep the first group.

Public Sector, Consumer Brands, Events and Workforce

The last four opportunities are less about one industry and more about how the whole region measures, spends and learns.

17. City and county services

City halls across the metro handle service requests, permits and infrastructure maintenance with limited staff. Agents can answer routine 311 questions, pre-check permit applications for missing items, and route requests to the right crew, while machine learning ranks which streets, pipes and buildings to fix first. Public spending also needs measurement: my open-source project Daedalus maps money flows and runs causal analysis on economic systems, the kind of tool that shows where a public dollar actually goes.

18. Marketing measurement for consumer brands

Kansas City headquarters consumer brands, Hallmark among them, that spend across retail, digital and seasonal campaigns. Bayesian marketing mix modeling with Papilon, validated against simulated data with PyCausalSim before anyone trusts it, replaces last-click attribution with estimates of what each dollar returned. Hallmark is one of the enterprises where I have worked, and the gap between what dashboards claim and what experiments show is usually large.

19. Measuring the World Cup’s real economic impact

Kansas City’s World Cup drew 68,000 to 69,000 fans to each of its four group-stage matches and more than 201,000 visits to the FIFA Fan Festival by the end of June, with a round of 32 match and a quarterfinal still to come in July (KCTV5). Before the tournament, KC2026, the local organizing committee, projected about 650,000 visitors and a $653 million economic impact (KCTV5), and local reporting asked how anyone would know how many actually came (HPPR). That is a causal inference question. Synthetic control and difference-in-differences methods, comparing Kansas City’s spending, hotel and tax data against similar metros that did not host, can estimate what the tournament added beyond what would have happened anyway. The same methods should be applied to every future stadium, convention and incentive decision the region makes.

20. AI workforce and small business adoption

The constraint on all nineteen opportunities above is people. A Heartland Forward poll found Kansans among the most optimistic about AI in the heartland, while small businesses still name cost, skills gaps and trust as their main barriers (Startland News). KC AI Lab showed one model that works: practitioners teaching practitioners on real data, at no cost to committed members. Scaling that model through employers, universities and community groups is the cheapest economic development lever on this list.

Where the Biggest Economic Impact Lands

Five areas stand out when the twenty are scored together, because they combine high decision volume, large value per decision, and reach beyond the metro.

1. Freight and logistics (opportunities 1 to 3). The highest decision volume in the region, with money attached to every load, door and pallet. Small percentage gains compound across millions of decisions a year.
2. Engineering and design (opportunity 10). Kansas City firms design infrastructure for the world, so a gain in engineering productivity here is exported in every project they deliver.
3. Grid and data center load (opportunity 11). Hundreds of megawatts of new demand make forecasting accuracy a regional cost issue, not just a utility one.
4. Animal health and biomanufacturing (opportunities 6 and 8). A corridor with a global share of its industry and new federal funding for biomanufacturing, where better evidence and process control raise the value of every product.
5. Workforce and small business adoption (opportunity 20). The broadest lever. It touches every other opportunity on the list and every company too small to hire a data science team.

The scale of the prize follows from the productivity numbers. At 2022 levels, Kansas City’s GDP per job trailed the peer benchmark by about $8,000. If decision systems raised output per job by even 1%, that would be about $990 per job per year at Kansas City’s 2022 level, spread across the whole regional workforce. That is an illustration, not a forecast, but it shows why decision quality, not chatbot adoption, is the measure that matters.

What Agentic Artificial Intelligence Means for Kansas City

Agentic artificial intelligence means AI systems that take actions inside business systems, not just answer questions: tendering a load, scheduling a crew, routing a claim, drafting a submission. Kansas City agentic artificial intelligence will look different from the Silicon Valley version, and it should.

This region’s companies run physical operations on tight margins, with systems of record they have trusted for decades. Kansas City agentic systems need to be built human-on-the-loop: people set the boundaries, economic limits and exception rules, agents execute the volume, and every action leaves an auditable record. They need causal measurement attached from day one, so leadership can see what each agent actually changed. And they need to be built on clean data, because an agent working from a stale vendor list makes the wrong decision at machine speed.

That is the work Vector1 Research was built for: agentic architecture with guardrails, causal measurement of what it delivers, and open-source tools that let a company’s own team check the math.

How a Kansas City Company Should Start

Pick one decision, not one technology. The companies that get value from AI in 2026 start small and measured:

1. Choose one high-volume decision from your own operation, scored with the formula above.
2. Measure the baseline against a comparison group before changing anything, so the result can be trusted.
3. Deploy with guardrails: clear limits, an exception path to a person, and a full record of every action.
4. Measure the effect causally, then expand to the next decision only when the first one has paid.

If you are working through any of these twenty opportunities in Kansas City, I want to compare notes. You can reach me at brian@vector1.ai, and the open-source tools are on GitHub.

Frequently Asked Questions

What are the biggest artificial intelligence opportunities in Kansas City?

The largest are in freight and logistics, engineering and design, grid and data center load forecasting, animal health and biomanufacturing, and workforce and small business adoption. They combine high decision volume, large value per decision and reach beyond the metro.

What is agentic artificial intelligence?

Agentic artificial intelligence refers to AI systems that take actions inside business systems, such as tendering freight, scheduling staff or routing claims, within limits set by people. In Kansas City, agentic systems are best built human-on-the-loop, with audit trails and causal measurement of their results.

How can causal inference help Kansas City businesses?

Causal inference measures what a decision actually changed, such as the effect of a price change, a promotion or a retention offer, by comparing against what would have happened without it. It replaces dashboards that credit whatever happened most recently.

Who is Brian Curry in Kansas City?

Brian James Curry is a Kansas City AI researcher and data scientist. He founded KC AI Lab, which began in 2016 as the Kansas City Machine Learning Group and grew to more than 1,500 members, and he founded Vector1 Research, an independent applied AI lab in Kansas City.

What is Vector1 Research?

Vector1 Research is a Kansas City applied AI lab founded by Brian James Curry. It works on agentic AI systems, causal inference, marketing economics and knowledge engineering, and publishes open-source tools including Papilon, PyCausalSim, MeaningFlow, Daedalus and tincast.

About the Author

Brian James Curry is a Kansas City AI researcher and data scientist, the founder of Vector1 Research and of KC AI Lab, which began in 2016 as the Kansas City Machine Learning Group and grew to more than 1,500 members. He has more than twenty years of experience leading analytics, data science and machine learning at enterprises including Hallmark, Garmin, Koch Industries, Tractor Supply Company and Vail Resorts. His open-source work, including Papilon, PyCausalSim, MeaningFlow, Daedalus and tincast, is on GitHub. He grew up in a small Kansas farming town and attended Kansas State University. He writes about causal inference, agent systems and the economics of AI on Medium and Substack. Contact: brian@vector1.ai.