How AI Is Helping Healthcare Companies in Buffalo Cut Costs and Improve Efficiency

New York's Empire AI investments are transforming Buffalo into an AI-for-health hub: recent state and philanthropic support is expanding UB's supercomputing capacity - most recently a $40 million award to build Empire AI Beta - so researchers can tackle medical imaging, drug discovery, mental‑health workforce shortages and hospital operations more effectively (University at Buffalo Empire AI $40M supercomputing award); UB's School of Public Health documents early clinical and population‑health pilots that use AI to detect disease earlier and optimize care pathways (UB School of Public Health AI and health research in Buffalo); and state leadership launched the consortium to advance AI for the public good (Governor Hochul Empire AI consortium announcement).

“With Empire AI, New York is leading in emerging technology and ensuring the power of AI is harnessed for public good.”

To turn this capacity into cost savings in Buffalo hospitals, local clinicians and administrators will need practical AI skills - training like Nucamp AI Essentials for Work bootcamp registration can help staff apply AI tools to documentation, billing and operational workflows.

Academic leadership in Buffalo is turning New York's Empire AI investment into hospital-ready tools by aligning university research, clinical partners, and local pilots.

University at Buffalo teams are building AI for motor-neuron disease diagnostics, improved medical imaging, and conversational tools that augment mental-health capacity (Empire AI biotech and drug discovery analysis), while state coordination and capital commitments expand shared supercomputing and research infrastructure that local teams can access (New York Empire AI consortium and state investment announcement).

These academic–clinical partnerships - from Roswell Park imaging workflow pilots to Kaleida Health documentation initiatives - create a pipeline for translational projects, student and faculty collaboration, and tech-transfer opportunities that reduce time-to-clinic and operational costs (Roswell Park GenAI imaging workflows and Buffalo AI healthcare use cases).

Together, this academic leadership supplies the models, compute, and trained workforce Buffalo hospitals need to scale AI projects that lower costs and improve care.

Buffalo hospitals can cut substantial administrative costs by deploying AI across revenue-cycle management, coding, and scheduling to reduce denials, speed reimbursements, and lower staffing overhead: national studies show AI-driven RCM and automation are becoming mainstream and can boost productivity and first-pass claim rates when paired with human oversight (AHA guide: AI for revenue-cycle management).

Smarter scheduling alone - predicting no-shows, automating reminders and waitlists - targets the $150 billion annual U.S. cost of missed appointments and the 25–30% no-show range while improving access and clinician utilization (CCD Health report: AI in healthcare scheduling and no-show statistics).

On billing and coding, AI claim-scrubbing, NLP-driven code suggestions, and automated appeal generation cut errors, shorten days in AR, and recover denied revenue - real-world vendors report large reductions in denials and faster cash collection when combined with bi-directional EHR integration and human-in-the-loop review (ENTER Health case study: medical billing automation and AI error reduction).

“cut down on documentation time and errors”

For Buffalo leaders, practical next steps are piloting targeted RCM and scheduling tools at Kaleida and Roswell Park, investing in staff AI upskilling, and enforcing governance so automation reduces cost without sacrificing compliance or equity.

Buffalo hospitals can boost clinical accuracy and patient safety by pairing locally developed AI imaging with proven decision‑support - early University at Buffalo work on OneTouch‑PAT (photoacoustic + ultrasound with deep‑learning reconstruction) produces AI‑enhanced 3D images in under a minute, visualizes tumor‑associated vasculature across subtypes, and reduces operator variability, offering a safer, less painful alternative for women with dense breasts (University at Buffalo OneTouch‑PAT breast imaging study results).

A practical companion is vendor AI that highlights suspicious findings and raises radiologist confidence - Johns Hopkins' rollout of AI decision‑support shows gains in reading efficiency while underscoring that algorithms are decision aids, not replacements, and must be validated against priors and bilateral comparisons (Johns Hopkins AI decision‑support improves breast imaging reading efficiency).

Local pilots can bridge research and the reading room by integrating UB tools, vendor software, and governance for safety and equity (Roswell Park GenAI imaging workflows for Buffalo radiology pilots).

“Our system, which is called OneTouch‑PAT, combines advanced imaging, automation and artificial intelligence - all while enhancing patient comfort.”

Practical next steps for Buffalo: run paired‑reading pilots, fund image‑quality validation on local cohorts, and train radiology teams in human‑in‑the‑loop workflows to realize accuracy and safety gains.

AI can drive measurable cost reductions across Buffalo's hospital supply chains, pharmacies, and therapeutics programs by combining demand forecasting, automated replenishment, and real‑time visibility: machine‑learning models reduce stockouts and overstocks, barcode/IoT counting and computer vision automate data capture, and anomaly detection flags unusual usage or spoilage to protect limited budgets and patient safety (see the practical hospital inventory AI methods in this hospital inventory AI and machine learning guide).

Applying these capabilities to Buffalo systems - starting with perioperative kits, high‑cost injectables, and short‑shelf‑life blood products at Kaleida Health and Roswell Park - lowers carrying costs, cuts expired‑item waste, and reduces emergency ad‑hoc transports that inflate logistics spend (see analysis of AI‑driven inventory optimization and patient safety).

A high‑impact case study shows how end‑to‑end ML forecasting and routing cut expiries and transport costs in a blood network; local pilots should mirror this phased approach, integrate EHR/ERP feeds, and keep human oversight for exceptions (see this AI platelet supply‑chain case study with results).

Key local metrics to track are stockout rate, days on hand, expiry rate and forecast accuracy:

“NHS Blood & Transplant are working with a company called Kortical to design an AI-powered supply and demand model for every hospital in England. It's cutting-edge tech supporting an age-old logistical challenge.”

Practical next steps for Buffalo: pilot pharmacy and blood‑product forecasting, invest in item master data quality, validate models on local consumption patterns, and train procurement and clinical staff for human‑in‑the‑loop workflows to secure savings without compromising care.

Buffalo hospitals can reduce revenue loss and recover mispaid funds by pairing machine‑learning fraud models with RPA for high‑volume claims verification and strict vendor governance: federal work for CMS shows AI in production surfaced more than $1 billion in suspect claims annually with >90% detection accuracy (CMS AI fraud detection case study showing $1B in suspect claims), and a 2025 systematic review finds ML techniques consistently improve claims‑fraud detection and explainability (Systematic review of machine‑learning techniques for healthcare claims fraud (2025)).

At the same time Buffalo faces real cybersecurity risk - 483,126 Catholic Health patient records were implicated in a 2024 vendor breach - so detection programs must be coupled with encryption, vendor SLAs, and rapid notification procedures (Serviceaide data breach impacting Catholic Health patients (2024)).

Key metrics to track locally are shown below, and practical steps include piloting ML+RPA claim‑scrubbing at Kaleida and Roswell Park, preserving human‑in‑the‑loop review for appeals, and sharing indicators with state partners to speed financial recovery.

“Humana's success demonstrates the value of AI in augmenting traditional fraud detection approaches. The system's capacity to learn and adapt has changed the game.”

Remote monitoring and telehealth are practical levers for Buffalo hospitals to lower costs and meet value‑based care goals by shifting care upstream, reducing avoidable readmissions, and enabling virtual chronic‑disease management that ties outcomes to payment.

Local pilots show how this works in practice: integrating device feeds and asynchronous imaging into clinician workflows can speed specialist review and triage (see Roswell Park's Oracle GenAI imaging workflows for Buffalo radiology), while Kaleida Health's early AI pilots tackling documentation and billing errors demonstrate the operational fixes needed to bill telehealth reliably and capture value under alternative payment models.

To scale these gains Buffalo must combine AI‑driven alert‑prioritization for remote patient monitoring with clear governance, EHR integration, and workforce upskilling so clinicians and care coordinators can adapt to role changes - guidance and training are especially important given research on the AI impact to Buffalo healthcare workers and strategies to adapt.

Practical next steps: pilot RPM programs with human‑in‑the‑loop triage, validate telehealth billing automation, and track readmission and total‑cost‑of‑care metrics to prove value for payors and patients.

Data governance, privacy, and equity must be built into every Buffalo AI pilot so cost‑saving tools don't amplify existing disparities: the University at Buffalo's new $3.6M NIH award to train equity‑focused researchers and embed community‑based participatory methods provides a practical foundation for governance that pairs technical controls with local accountability (University at Buffalo $3.6M NIH grant for health inequities and Center of Excellence).

Local imaging and operational pilots should follow with clear data‑use agreements, vendor SLAs, encryption and breach plans, bias audits, and representative local datasets - steps already considered in Roswell Park GenAI workflow pilots that accelerate diagnostic review while raising governance questions (Roswell Park GenAI imaging workflow pilots and governance considerations) and Kaleida Health documentation pilots that show how governance and human‑in‑the‑loop review preserve safety and reimbursement integrity (Kaleida Health AI documentation pilot guide and governance practices).

Prioritize community engagement, transparent consent language, ongoing monitoring of equity metrics, and workforce training so gains in efficiency translate into fairer outcomes for Buffalo patients.

“The work of health equity has got to be data-driven... When we better understand the scope of the problems and what's driving them, then we can develop approaches and remedies to solve them.”

AI investment is already reshaping Buffalo's community and jobs: local ecosystem grants and competitions are seeding organizations that train entrepreneurs, connect founders to mentors, and channel AI and biotech projects into the region - most recently Springboard awarded $551,625 across 12 ecosystem builders to accelerate startups and inclusive supports (Springboard awards $551,625 to Buffalo entrepreneurial ecosystem).

Venture development groups are translating that momentum into healthcare-focused startups and hires: Launch NY's $200K investment in PhysicianX is one example of targeted capital that helps retain clinical talent, scale AI tools for physicians, and build user-growth that becomes local jobs and services (Launch NY invests $200K in PhysicianX to support Buffalo startups).

At the macro level, New York's Empire AI build‑out expands UB-hosted supercomputing and SUNY access - public/private commitments now exceed hundreds of millions and will attract research, internships, and industry partners that supply the talent pipeline Buffalo hospitals need (Empire AI expansion and SUNY AI capability investments announced by Governor's Office).

“For any region within the country, you really need entrepreneurship to be the backbone of growth…that's how you keep your young people in town.”

Practical implications for Buffalo healthcare: prioritize local hiring pipelines, fund AI upskilling and clinical internships, and align procurement/pilots so AI-driven efficiency translates into sustainable jobs and broader economic gains.

Practical next steps for Buffalo healthcare organizations are to adopt a phased, measurable approach: run focused pilots for revenue-cycle claim-scrubbing, intelligent scheduling to cut no-shows, and inventory forecasting for high-cost items; build vendor SLAs, bias audits, encryption and human-in-the-loop review into every pilot; and track clear KPIs so savings scale safely.

Start with proven administrative automations - scheduling, billing, and EHR documentation workflows - to free clinician time and reduce errors (see the Keragon AI in Healthcare Administration guide: Keragon AI in Healthcare Administration guide), and mirror operational pilots that show tangible workflow gains (evidence on AI hospital workflow automation: Tipstat AI hospital workflow automation evidence).

Train staff in prompt-engineering and human-centered AI workflows so clinicians and administrators retain oversight - practical upskilling like the Nucamp AI Essentials for Work bootcamp helps nontechnical staff apply tools safely (register for the Nucamp AI Essentials for Work bootcamp: Nucamp AI Essentials for Work bootcamp registration).

“AI and generative AI hold promise to enhance diagnostic accuracy, improve workflow efficiency, and advance education and research. Integration of AI must be thoughtful, addressing ethics and validation to ensure patient care standards.”

Prioritize governance, iterative validation on local data, and partnerships with UB, Roswell and Kaleida to turn pilots into sustained cost reductions without sacrificing equity or safety.