Category: 3. Business

Merger of Equals with Anglo American plc approved by both classes of Teck shareholders

Vancouver, B.C. – Teck Resources Limited (TSX: TECK.A and TECK.B, NYSE: TECK) (“Teck”) announced today the voting results from its Special Meeting of Shareholders held on Tuesday, December 9, 2025 (the “Meeting”). Teck shareholders overwhelmingly voted to approve the special resolution (the “Arrangement Resolution”) approving the plan of arrangement under the Canadian Business Corporations Act, involving, among other things, the “merger of equals” of Anglo American plc (“Anglo American”) and Teck (the “Merger”). 99.7% of the votes cast by Class A common shareholders at the Meeting were in favour of the Arrangement Resolution and 89.7% of votes cast by Class B subordinate voting shareholders were in favour of the Arrangement Resolution. 

“This resoundingly positive vote marks an important milestone in creating Anglo Teck—a global leader in critical minerals headquartered in Canada,” said Jonathan Price, President and CEO, Teck. “Anglo Teck will be positioned to deliver long-term value through a world-class copper growth portfolio, operational and functional synergies, and a stronger platform to meet growing demand for critical minerals essential to global economic growth and the energy transition. We look forward to advancing the necessary regulatory approvals and completing the merger for the benefit of our shareholders, employees, communities, and partners.”

The Arrangement Resolution required the approval of at least (a) two-thirds of the votes cast by Class A common shareholders present or represented by proxy at the Meeting, voting separately as a class; and (b) two-thirds of the votes cast by Class B subordinate voting shareholders present or represented by proxy at the Meeting, voting separately as a class. A total of 6,329,767 Class A common shares, representing 83.3% of the votes attached to all outstanding Class A common shares, and 380,842,347 Class B subordinate voting shares, representing 79.4% of the votes attached to all outstanding shares, were voted at the Meeting.  Detailed voting results for the Meeting will be available under Teck’s profiles on SEDAR+ (www.sedarplus.ca) and EDGAR (www.sec.gov).

As announced earlier today by Anglo American, at the General Meeting of Anglo American held earlier on December 9, 2025, shareholders of Anglo American overwhelmingly approved (i) the allotment and issue of new ordinary shares of Anglo American to Teck shareholders in connection with the Merger; and (ii) the change of the name of Anglo American to “Anglo Teck plc” with effect from the completion of the Merger.

The Merger remains subject to customary closing conditions, including approval under the Investment Canada Act and applicable competition and regulatory approvals in various jurisdictions globally and final approval by the Supreme Court of British Columbia. Further information about the Merger can be found in Teck’s management information circular dated November 3, 2025 (the “Circular”) for the Meeting, which is available under Teck’s profile on SEDAR+ and on EDGAR.  As disclosed in the Circular, prior to the deadline for eligible Canadian Teck shareholders to make an election to receive the exchangeable share consideration under the Merger, Teck will provide to registered Teck shareholders a letter of transmittal and election form (the “Letter of Transmittal and Election Form”). The Letter of Transmittal and Election Form will explain how to exchange Teck shares for the consideration under the Merger and, for eligible Canadian Teck shareholders, how to elect to receive the exchangeable share consideration under the Merger. Teck will issue a news release announcing once the Letter of Transmittal and Election Form has been made available and providing details on the relevant exchange and election procedures.

Forward Looking Statements
This news release contains certain forward-looking information and forward-looking statements as defined in applicable securities laws (collectively referred to as forward-looking statements). These statements relate to future events or future performance. All statements other than statements of historical fact are forward-looking statements. The use of any of the words “anticipate”, “can”, “could”, “plan”, “continue”, “estimate”, “expect”, “may”, “will”, “would”, “project”, “predict”, “likely”, “potential”, “should”, “believe” and similar expressions is intended to identify forward-looking statements. These statements involve known and unknown risks, uncertainties and other factors that may cause actual results or events to differ materially from those anticipated in such forward-looking statements. These statements speak only as of the date of this news release. These forward-looking statements include, but are not limited to, statements concerning the anticipated benefits and synergies from the proposed Merger, the expected effects of the Merger on Anglo American and Teck, future production levels, the expected timing of completion of the Merger, and other statements that are not historical facts.

These statements are based on a number of assumptions, including, but not limited to, assumptions regarding general business and economic conditions, future outlook and anticipated events, such as the ability of Anglo American and Teck to complete the Merger, the ability of Teck and Anglo American to obtain all required regulatory and court approvals, the ability of Teck and Anglo American to satisfy all other conditions to the Merger, the strategic vision of the merger between Teck and Anglo American following the closing of the Merger, expectations regarding exploration, production and operational potential, expectations with respect to production capabilities and future financial or operating performance of Teck and Anglo American following the Merger, expectations with respect to Teck’s current production and cost guidance and previously disclosed updates, the potential valuation of the merger of Teck and Anglo American, the expected synergies between Teck and Anglo American, the expected revenue from the synergies between Teck and Anglo American, expectations regarding integration and synergy capture; the accuracy of the pro forma financial position and outlook of Teck and Anglo American following the closing of the Merger, the success of the new board and management team, the satisfaction of the conditions precedent to the Merger, the future financial or operating performance of the merged Teck and Anglo American, the expected EBITDA uplift, the expectations around the headquarters of the combined entity being in Canada, the expectations of the results and success of the Investment Canada Act commitments, the expectations with respect to receiving Investment Canada Act approval, the assumptions surrounding the proposed Investment Canada Act commitments, the expectations with respect to the proposed investments by the combined company in Canada, the potential of Teck and Anglo American following the Merger to meet industry target, public profile expectations, future plans, projections, objectives, estimates and forecasts and the timing related thereto and the expectations surrounding the combined companies long-term strategy. The foregoing list of assumptions is not exhaustive. Events or circumstances could cause actual results to vary materially.

Forward-looking information is based on the information available at the time those statements are made and are of good faith belief of the officers and directors of Teck and Anglo American as of the time with respect to future events and are subject to risks and uncertainties that could cause actual results to differ materially from those expressed in the Forward-looking information. Factors that may cause actual results to vary materially include, but are not limited to, the possibility that the Merger will not be completed on the terms and conditions, or on the timing, currently contemplated, and that it may not be completed at all, due to a failure to obtain or satisfy, in a timely manner or otherwise, required regulatory and court approvals and other conditions to the closing of the Merger or for other reasons, public perception of the Merger, market reaction to the Merger, the negative impact that the failure to complete the Merger for any reason could have on the business of Anglo American or Teck, the ability of Anglo American and Teck to successfully integrate and capture expected synergies, general economic and market conditions, including interest and foreign exchange rates, global financial markets, changes in government regulations or in tax laws, industry competition, technological developments and other factors described or discussed in Anglo American’s or Teck’s disclosure materials filed with applicable securities regulatory authorities from time to time.
Teck assumes no obligation to update forward-looking statements except as required under securities laws. Further information concerning risks, assumptions and uncertainties associated with these forward-looking statements, the Merger and Teck’s business can be found in Teck’s Circular in respect of the Meeting filed under Teck’s profile on SEDAR+ (www.sedarplus.ca) and on EDGAR (www.sec.gov).

About Teck
Teck is a leading Canadian resource company focused on responsibly providing metals essential to economic development and the energy transition. Teck has a portfolio of world-class copper and zinc operations across North and South America and an industry-leading copper growth pipeline. We are focused on creating value by advancing responsible growth and ensuring resilience built on a foundation of stakeholder trust. Headquartered in Vancouver, Canada, Teck’s shares are listed on the Toronto Stock Exchange under the symbols TECK.A and TECK.B and the New York Stock Exchange under the symbol TECK. Learn more about Teck at www.teck.com or follow @TeckResources.

Investor Contact:
Emma Chapman
Vice President, Investor Relations
+44.207.509.6576
emma.chapman@teck.com

Media Contact:
Dale Steeves
Director, External Communications
236.987.7405
dale.steeves@teck.com

25-33-TR

Key Facts

Big Number

Nearly 109%. That’s how much spot silver has increased this year, outpacing gold, which has surged 60% while setting several milestones. Spot gold has increased just 0.4% over the last day to around $4,226, after hitting an all-time high above $4,381 in October. Spot platinum has also outpaced gold, rallying 86% as demand for electric vehicles has lifted platinum’s value in recent years, while global supply declines.

Read More

How we did this

Pew Research Center conducted this study to better understand teens’ use of social media, the internet and artificial intelligence (AI) chatbots.

The Center conducted an online survey of 1,458 U.S. teens from Sept. 25 to Oct. 9, 2025, through Ipsos. Ipsos recruited the teens via their parents, who were part of its KnowledgePanel. The KnowledgePanel is a probability-based web panel recruited primarily through national, random sampling of residential addresses. The survey was weighted to be representative of U.S. teens ages 13 to 17 who live with their parents by age, gender, race and ethnicity, household income, and other categories.

Here are the questions used for this report, along with responses, and the survey methodology­­­.

This research was reviewed and approved by an external institutional review board (IRB), Advarra, an independent committee of experts specializing in helping to protect the rights of research participants.

Even as teens express mixed feelings about social media’s impact, these sites remain a key part of their lives, with some using them “almost constantly.”

Now, AI chatbots, like ChatGPT and Character.ai, are getting teens’ attention. Roughly two-thirds report using chatbots, including about three-in-ten who do so daily, according to a new Pew Research Center survey of 1,458 U.S. teens ages 13 to 17.

Which online platforms teens use

Young people turn to a variety of platforms, but YouTube stands out for being used by nearly all teens. Roughly nine-in-ten report ever using it.

Teens widely use three other platforms:

• About six-in-ten or more say they use TikTok and Instagram.
• A somewhat smaller share say they go on Snapchat (55%).

Fewer use Facebook (31%) and WhatsApp (24%). And no more than about one-in-five say the same of Reddit or X (formerly Twitter).

Changes over time

Today’s online landscape for teens is marked by both stability and new trends.

WhatsApp is one platform that stands out for its growth in recent years. Today, roughly a quarter of teens say they use WhatsApp, up from 17% in 2022.

X and Facebook have declined in use over the past decade. Today, 16% of teens use X, down from 23% in 2022 and 33% in 2014-15. And Facebook, once the go-to platform for teens, is used today by about three-in-ten teens. This is far lower than the 71% in 2014-15, though on par with 2022.

The shares of teens who use other sites or apps, like YouTube, TikTok and Instagram, have stayed relatively stable in recent years.

Jump to read about teens’ online experiences: Online platform use by demographic groups | Frequency of online platform use | Use of AI chatbots | Frequency of chatbot use | Internet use

Online platform use by demographic groups

Teen use of specific online platforms varies across demographic groups – including when it comes to gender, race and ethnicity, age and household income.

By gender

Teen girls are more likely to use Snapchat and Instagram. For example, 61% of girls say they use Snapchat, compared with 49% of boys.

Meanwhile, boys are more likely to use Reddit (21% vs. 12%) and YouTube (94% vs. 89%).

By race and ethnicity

There are differences in use by race and ethnicity across all the platforms asked about except Reddit. Black teens are more likely than their White or Hispanic peers to use Instagram, TikTok, X, Snapchat and YouTube. For example, 82% of Black teens say they use Instagram. This drops to 69% among Hispanic teens and is even lower for White teens (55%). And Black teens are more likely than Hispanic teens to use Facebook.

WhatsApp is used by a larger share of Hispanic and Black teens than White teens.

By age

Older teens stand out from younger teens in using nearly every platform we ask about. For instance, three-quarters of 15- to 17-year-olds say they use Instagram, compared with 44% of 13- to 14-year-olds.

YouTube is the only site measured that older and younger teens are equally likely to use.

By household income

Teens in households with lower and middle incomes are more commonly using TikTok and Facebook, a largely similar pattern to previous years.

For instance, 46% of teens living in households earning less than $30,000 a year say they use Facebook. Similarly, 39% of those in households with incomes between $30,000 and $74,999 say the same. However, this drops to 27% among teens in households earning $75,000 or more.

By party

In a pattern seen in previous Center surveys, a larger share of teens who identify as Democrats than Republicans say they use TikTok, Instagram, Reddit and YouTube.

For example, there is a large partisan gap for TikTok: 75% of Democratic and Democratic-leaning teens say they use TikTok, compared with 60% of Republicans and Republican leaners.

Frequency of online platform use

YouTube is not only widely used, but it’s also the platform the most teens visit on a daily basis. Roughly three-quarters of teens say they use it every day.

Somewhat smaller shares report going on two other platforms daily: TikTok (61%) and Instagram (55%).

Just under half say they visit Snapchat every day (46%), while far fewer say the same of Facebook (20%).

Overall, teen daily use of these platforms remains relatively stable from past years.

Social media is not only a daily feature in the lives of teens, some report using these platforms “almost constantly.” About one-in-five teens say this of TikTok and YouTube.

Fewer describe their use of Instagram and Snapchat as almost constant (12% for each). And just 3% say this of Facebook.

Across these five platforms, 36% of teens use at least one of these sites almost constantly.

Changes over time

The share of teens who say they are on TikTok almost constantly ticked up slightly to 21% this year, from 16% in 2022. The shares who report using YouTube, Instagram, Snapchat and Facebook almost constantly have changed little since 2022.

By gender

There are some gender differences in frequency of using these sites or apps.

Slightly larger shares of teen girls than boys report being on TikTok and Instagram almost constantly. Teen boys are more likely than girls to visit YouTube this often (20% vs. 13%).

Similar rates of girls and boys say they use Snapchat and Facebook almost constantly.

By race and ethnicity

Black and Hispanic teens are particularly likely to report being on TikTok, YouTube and Instagram almost constantly.

For example, 35% of Black teens say they’re on YouTube almost constantly, compared with 23% among Hispanic teens. Both groups are much more likely than White teens (8%) to say this.

There are only small or no racial or ethnic differences in visiting Snapchat or Facebook almost constantly.

Use of AI chatbots

AI chatbots have become more common in daily life, from education to entertainment. For the first time, we asked teens about their overall use of chatbots, how often they use them and which ones they turn to.

A majority of teens say they use chatbots. Roughly two-thirds of teens (64%) say they ever use an AI chatbot. Fewer (36%) do not use this tool.

While many teens use chatbots, there are some differences across demographic groups:

• Race and ethnicity: Roughly seven-in-ten Black and Hispanic teens say they use chatbots, higher than among White teens (58%).
• Age: 68% of teens ages 15 to 17 use chatbots, compared with 57% among teens 13 to 14 years old.
• Household income: Teens living in households earning $75,000 or more are more likely than those in households with incomes of less than $30,000 to use chatbots (66% vs. 56%). Those living in households earning $30,000 to $74,999 do not differ from either group.

Frequency of chatbot use

About three-in-ten teens say they use AI chatbots every day, including 16% who do so several times a day or almost constantly.

Daily use of chatbots differs somewhat by race and ethnicity as well as age:

• Race and ethnicity: About a third of Black (35%) and Hispanic teens (33%) report using AI chatbots daily. A smaller share of White teens (22%) say the same.
• Age: 31% of teens ages 15 to 17 say they use chatbots on a daily basis, compared with about a quarter of those ages 13 to 14 (24%).

Which chatbots do teens use?

In addition to understanding their overall use, we also asked teens about their use of six specific chatbots.

ChatGPT (59%) is by far the most widely used chatbot and the only one we measured that a majority of teens use.

This is more than twice the rate of the next most commonly used chatbots: Gemini (23%) and Meta AI (20%).

Fewer say they use Copilot, Character.ai and Claude.

By race and ethnicity

Black and Hispanic teens are more likely than their White peers to say they use Gemini and Meta AI.

Black and White teens differ modestly in their use of ChatGPT and Character.ai.

There are no significant differences in use for Copilot or Claude.

By age

Teens ages 15 to 17 are more likely than those 13 to 15 to report using ChatGPT and Meta AI.

By household income

ChatGPT use is more common among teens in higher-income households. About six-in-ten teens living in households earning $75,000 or more (62%) say they use it. That compares with 52% of teens living in households earning less than $75,000.

Meanwhile, lower- and middle-income teens are more likely to use Character.ai. Some 14% of teens in households with incomes of less than $75,000 report using it. This is double the rate among teens in households with incomes of $75,000 or more (7%).

Go to the appendix for a full breakdown of AI chatbot use by demographic groups.

Teens’ internet use

The survey also explores how often teens use the internet.

Nearly all U.S. teens (97%) say they use the internet daily, including four-in-ten who say they are almost constantly online.

The share of teens who say they’re online almost constantly is much higher today than a decade ago, though it’s a slight dip from last year.

By race and ethnicity

Black (55%) and Hispanic teens (52%) are about twice as likely as White teens (27%) to say they’re online almost constantly.

By age

Being online almost constantly is more common for older teens. While 43% of 15- to 17-year-olds report being online almost constantly, 34% of 13- and 14-year-olds report this.

By household income

Teens living in households that earn less than $75,000 annually are more likely than those in households earning $75,000 or more to say they use the internet almost constantly.

There are no significant differences in internet use by gender.

Introduction

Machine learning (ML) has significant potential to enhance diagnostic accuracy, support clinical decision-making, and facilitate early disease detection [-]. Additionally, contemporary ML technologies that streamline administrative tasks, refine billing and coding workflows, and enhance patient flow may substantially benefit hospital operations, particularly in underserved areas facing workforce shortages [-]. As recent ML advancements shift the focus from development to deployment, US health care providers have increasingly embraced these technologies [-], especially through integration with electronic health records (EHRs) [-]. Integrating ML techniques into EHRs offers multiple advantages such as simplifying data sharing, reducing analytical time, and mitigating the risk of data leakage []. These techniques enable the efficient processing and analysis of vast amounts of unstructured EHR data, which is critical for research but remains largely underutilized []. Additionally, ML-driven automation can help alleviate provider burnout by reducing manual data entry, minimizing human errors, and shortening documentation time [,].

Despite significant benefits, several technical and ethical challenges (eg, patient privacy, model accuracy, and data reliability) associated with ML adoption may result in unanticipated adverse consequences for patients [,-]. Moreover, the implementation of new health technology is often complicated by various social and organizational factors, posing potential risks to care delivery [,]. For instance, hospitals serving more vulnerable populations may lack adequate technical support or sufficient staff when implementing ML in EHRs []. Thus, both the benefits and challenges warrant further exploration of the factors influencing ML adoption, as hospitals continue to enhance clinical outcomes and operational efficiency by embracing new health technologies.

Although prior work has examined the implementation of ML in hospital settings [-], most studies concentrated on overall adoption, typically measured as the presence of any ML functions or the number of ML functions adopted in EHRs. Such an oversimplified classification treats ML-enabled tools as a homogeneous technology and obscures substantial variation across ML functions. For instance, ML for patient scheduling and ML for inpatient risk prediction differ markedly in data requirements, implementation complexity, and regulatory oversight [-]. Prior studies may mask meaningful heterogeneity in real-world ML implementation. Moreover, hospital adoption patterns may vary across particular ML domains or functions depending on distinct organizational needs, resource capacities, and technological benefits (eg, a large academic hospital may prioritize ML for clinical decision support, while a smaller facility may favor administrative automation). Prior research focusing solely on overall adoption may fail to capture this nuance. To address these gaps and better reflect the complexity of ML diffusion in real-world hospitals, our study differentiates 3 dimensions of adoption: (1) overall ML adoption, (2) domain-specific ML adoption at different levels, and (3) adoption of individual ML functions. This multidimensional approach yields a more granular understanding of how hospitals implement ML within EHRs and how adoption patterns differ by institutional context.

In this study, our goals are to (1) describe the status quo of ML adoption in EHRs across US general acute hospitals and (2) identify hospitals and characteristics associated with ML adoption in EHRs, guided by the Technology-Organization-Environment (TOE) framework. Leveraging a national hospital sample and regression analysis, our study offers valuable insights for policymakers, health care administrators, and clinicians regarding the current landscape of ML adoption and for identifying strategies to navigate the opportunities and challenges posed by emerging technologies.

Methods

Conceptual Framework

In this study, we integrated the TOE framework with our regression analysis to explain the characteristics that are associated with the ML adoption in EHRs [,]. TOE suggests that ML adoption decisions are shaped by 3 contexts: technology, organization, and external environment. The technological context reflects technical benefits and costs of the current ML innovation, such as ML relative advantage, implementation complexity, compatibility with existing EHR systems and workflows, and technical maturity. The organizational context captures hospital internal readiness and capability, including hospital size and structure, leadership support, IT staffing, available financial resources, and IT-supportive organizational culture. Environmental context encompasses all external enablers or barriers outside hospitals such as market competition, artificial intelligence (AI)–related regulations and policy, or EHR vendor support. In health care settings, TOE has been widely used to explain health technology adoption in hospitals [-]. These characteristics in TOE offer inevitable clues of understanding who adopts, who benefits, and where digital capability gaps persist. Recognizing these factors in the adoption of ML technology is critical for designing future policies and practice interventions that promote equitable and effective ML implementation across the US health care system.

Data and Sample

We utilized data from the 2022‐2023 American Hospital Association (AHA) Annual Survey and the 2023‐2024 AHA IT Supplement Survey to examine the relationship between hospital characteristics and ML adoption into EHR []. Each year, the AHA surveys health care administrators from 6200 US hospitals and 400 health care systems, collecting information on hospital demographics, operations, and financial metrics. In addition to the main survey, the AHA conducts the IT Supplement Survey, which gathers data on health IT adoption, tools, and barriers across several domains, including patient engagement, social determinants of health, health information exchange, EHR systems, and IT vendors. The IT Supplement Survey is typically completed by the chief information officer, and participation is voluntary. Beginning in 2023, the IT Supplement introduced a series of questions on ML and other predictive models, enabling the investigation of various aspects of ML adoption, such as functionality, developers, and model evaluation practices.

Our study cohort comprises US general and acute care hospitals operating across all 50 states and the District of Columbia that successfully responded to the 2023‐2024 AHA IT Supplement Survey. After applying pairwise deletion for missing data and implementing other exclusion criteria, the final analytic sample includes 2562 unique hospitals across multiple years, resulting in a total of 4055 hospital-year observations.

ML Measures

ML adoption is measured based on 3 sets of outcomes using the survey questions from the AHA IT Supplement. The first outcome is a binary indicator for any ML adoption, which equals 1 if a hospital uses any ML or other predictive models. Second, to capture the specific uses of this ML, we used a follow-up question to create a 4-category measure. This question asked respondents to select from a list of specific ML applications, which we grouped into 2 domains. Clinical functions included (1) predicting health trajectories or risks for inpatients, such as early detection of conditions like sepsis or in-hospital fall risk; (2) identifying high-risk outpatients to inform follow-up care (eg, readmission risk); (3) monitoring health through integration with wearable devices; and (4) recommending treatments by identifying similar patients and their outcomes. Operational functions included (5) simplifying or automating billing procedures and (6) facilitating scheduling, such as predicting no-shows or optimizing block utilization. Based on both domains, we classified hospitals’ ML use into 1 of 4 mutually exclusive categories: (1) no ML adoption, (2) adoption of clinical functions only, (3) adoption of operational functions only, and (4) adoption of both. This categorical outcome allows for a more nuanced analysis of a hospital’s implementation strategy and provides an interpretable proxy for its organizational readiness. Open-ended responses for other functions were excluded due to their low frequency. Finally, we also analyzed the adoption of individual ML functions listed above.

Hospital and Environmental Factors

Inspired by the TOE framework and prior studies of health IT adoption [], we selected the following hospital characteristics as explanatory variables. For organizational context, we included hospital ownership (defined as nonfederal governmental, not-for-profit, and for-profit), bed size (classified as small [0‐99 beds], medium [100‐399 beds], and large [400 or more beds]), critical access hospital (CAH) status, and teaching hospital status. Regarding environmental context, we included health system affiliation (defined as whether hospitals are affiliated with health systems), whether to contract with leading EHR vendors which is identified by EHR market share [], and metropolitan location (ie, if a hospital is located in a county identified by Rural-Urban Continuum Codes Category 1‐3, it is defined as a metropolitan hospital, otherwise a nonmetropolitan hospital). Due to the lack of relevant technical context information in the AHA survey, this study did not include the technological context.

Statistical Analysis

First, we reported descriptive statistics for ML adoption, specific ML functions adopted in EHR, who developed ML in EHR, and how hospitals evaluated ML models (ie, accuracy, bias, and postimplementation). Second, we summarized ML adoption status by different hospital characteristics. Chi-squared tests were performed to determine statistically significant differences between hospitals with and without ML adoption.

We further assessed the associations between hospital characteristics and ML adoption using multivariate regressions. We used logistic regressions for ML adoption and multinomial logistic regression for a 4-category measure of ML adoption type. Besides explanatory variables listed above, all regressions controlled for year and geographic region based on Census Divisions (ie, New England, Middle Atlantic, East North Central, West North Central, South Atlantic, East South Central, West South Central, Mountain, and Pacific). For all outcomes, we report marginal effects with 95% CIs, which can be directly interpreted as changes in percentage points in the likelihood for a specific category of ML adoption [,]. SEs are clustered at the hospital level to account for a within-hospital correlation over time. We consider estimates to be statistically significant at P<.05. All statistical analyses are performed using StataNow/SE 18.5 version [].

Given high nonresponse rates in the AHA IT Supplement, we compared the characteristics of respondents versus nonrespondents and found that nonrespondent hospitals were more likely to be smaller, less system affiliated, and located in rural areas—factors which are potentially associated with lower adoption rates (Table S1 in ). To address potential nonresponse bias and produce nationally representative estimates, we constructed inverse probability weights (IPWs) from the regression that estimated each hospital’s probability of responding to the AHA IT Supplement (propensity score) using the same set of characteristics described above as predictors [,]. IPW mitigates selection bias by reweighting observations based on their propensity to respond to the AHA IT Supplement, thereby creating a pseudo-population where the distribution of observed covariates is balanced between respondents and nonrespondents. Specifically, for each year, we fitted a separate logistic regression model and predicted the probability of responding to the AHA IT Supplement. We then constructed the analysis weights as the inverse of this predicted response probability and applied these weights in all our statistical analyses. After adjusting by IPWs, the previously observed discrepancies between respondents and nonrespondents became smaller (Table S1 in ), which suggests mitigating the nonresponse bias. All descriptive analyses and regressions are weighted using the IPWs. It is important to note that this IPW approach assumes that nonresponse depends only on observables; accordingly, it corrects for bias related to the measured hospital characteristics. However, any unobserved factors correlated with both survey nonresponse and ML adoption could still bias our estimates, which could not be fully addressed in our study.

Ethical Considerations

Our study is a secondary analysis of organization-level data whose original survey design and administration are described in AHA technical documentation [], and does not involve human participants. Under the US Common Rule (45 CFR §46), research using deidentified, organization-level data does not constitute human subjects research and therefore does not require Institutional Review Board oversight []. We accessed the AHA data via Wharton Research Data Services (WRDS), under data sharing agreements between the AHA and WRDS and between WRDS and the University of Alabama at Birmingham. The results are reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.

Results

Descriptive Analysis

Overall ML Adoption

The flowchart of our sample selection is presented in Figure S1 in . Among our 2023‐2024 analytical sample, 73%‐76% of the hospitals reported any ML functionality within their EHR system (Figure S2 in ). The majority of US hospitals tend to adopt both clinical and operative ML functions, while the share increased by 10.7 percentage points from 2023 to 2024.

As summarized in , 83.8% of the metropolitan hospitals adopted ML in EHRs, while only 61.3% of the nonmetropolitan hospitals adopted MLs (all subsequent percentages follow the same interpretation). Adoption rates were higher among large hospitals (94.4% vs 65.0% for small), those contracted with leading EHR vendors (92.1% vs 56.7% for nonleading EHR), not-for-profit hospitals (82.7% vs 69.8% for for-profit), non-CAHs (82.2% vs 57.4% for CAHs), those affiliated with health systems (87.9% vs 40.1% for non-health-system), and teaching hospitals (92.1% vs 56.7% for non-teaching). The geographic distribution of any ML adoption is exhibited in Figure S3 in .

Specific ML Function

The most widely adopted functions (Figure S4 in ) are predicting health trajectories or inpatient risks and identifying high-risk outpatients for follow-up care. From 2023 to 2024, the adoption of clinical ML functions remained stable, whereas operational functions rose markedly, specifically by 19.9 percentage points for simplifying or automating billing procedures and by 14.3 percentage points for facilitating scheduling.

ML Development

Hospitals leveraged multiple resources to develop ML or other predictive models (Figure S5 in ). About 70%‐80% of hospitals reported that their ML was developed by their EHR vendors. The third-party developers and in-house IT teams continued to play meaningful roles in ML development, which likely reflects hospitals’ needs in customizing features to meet varying clinical and administrative needs. Notably, the share of hospitals unsure of the source rose sharply from 1.0% in 2023 to 16.8% in 2024.

ML Model Evaluation

Among hospitals adopting ML in 2023, 62.6% reported evaluating model accuracy for all or most models, while only 45% assessed model bias (Figure S6 in ). These rates of model accuracy and bias evaluation increased to 70.7% and 56.9% in 2024. In 2024, 58.1% of the hospitals conducted postimplementation evaluation and monitoring.

Regression Analysis Results 

Whether to Adopt Any ML

The IPW weighted regression estimates of the associations between ML adoption and hospital characteristics are reported in . Compared to nonfederal governmental hospitals, not-for-profit hospitals were 4.4 (95% CI 0.6-8.2) percentage points more likely to adopt any ML within EHR systems (P=.02), while for-profit hospitals were 8.5 (95% CI −15 to −2.1) percentage points (P=.009) less likely to adopt. Large hospitals were 15.2 (95% CI 9.4-21) percentage points more likely than small hospitals to adopt ML into EHRs (P<.001). Hospitals affiliated with a health system had a 26.8 (95% CI 22.4-31.3) percentage point higher likelihood of ML adoption (P<.001). In contrast, CAHs were 8.4 (95% CI –12.4 to –4.4) percentage points less likely to adopt ML (P<.001). Contracting with leading EHR vendors was associated with a 20.6 (95% CI 17.1-24) percentage points increase in ML adoption likelihood (P<.001). Hospitals located in metropolitan areas were 4.3 (95% CI 0.8-7.8) percentage points more likely to adopt ML technology compared to those in nonmetropolitan areas (P=.02).

Type of ML Adoption

The multinomial logistic regression results in illustrate the association between hospital characteristics and 4 types of ML adoption. The results of both clinical and operational ML adoption are overall consistent with the estimates on any ML adoption: not-for-profit, large size, health system affiliation, non-CAH, leading EHR contract, and metropolitan location were related to both ML adoption. Regarding other types of ML adoption, for-profit (−15.1; 95% CI −20 to −10.2; P<.001) and teaching (−6.2; 95% CI −11 to −1.4; P<.001) hospitals were less likely to adopt only clinical ML. Large size (−1; 95% CI −2 to −0.1; P=.03) and health system affiliation (−1.8; 95% CI −3.4 to −0.2; P=.03) were related to a lower likelihood of only operational ML adoption.

Individual ML Functions

We further explored the relationship between hospital features and 6 individual ML functions in Tables S2 and S3 in . Overall, the patterns were consistent with the main findings in , but some exceptions emerged. Nonprofit hospitals showed a stronger emphasis on clinical applications, including predicting health risks for inpatients (7.5; 95% CI 3.6-11.4; P<.001) and outpatients (7.7; 95% CI 3.4-12; P<.001). In contrast, for-profit hospitals (26.4; 95% CI 19.3-33.6; P<.001) and large hospitals (11.6; 95% CI 5.3-18; P<.001) were more likely to adopt scheduling-related ML functions. To enable clearer review and a comparison of evidence across various sets of outcomes, we provided a summary overview in .

Additional Analysis Results 

To check the robustness of our findings, we estimated models without IPWs and compared them with our main estimates (Tables S4 and S5 in ). We found both weighted and nonweighted estimates converge. The estimates are consistent with prior research on hospital AI adoption for workforce optimization [,], lending credibility to our findings.

 We also conducted the subgroup analysis to detect heterogeneity patterns in different hospital sizes (Table S6 in ). Health system affiliation and leading EHR vendors significantly increased the probability of ML adoption, and both effects are larger in small hospitals than in medium hospitals.

Discussion

Principal Results

This study provides timely, national-level evidence on ML adoption within hospital EHR systems, revealing substantial variation in specific functions, model evaluation practices, and hospital characteristics. Consistent with prior research on overall ML adoption, organizational and environmental factors such as hospital type, size, CAH status, health system membership, and metropolitan location remain key determinants of adoption decisions, supporting the organizational and environmental contexts in the TOE framework. However, our findings extend beyond existing studies by identifying marked heterogeneity in the adoption of individual ML domains and functions, highlighting that hospitals differ not only in whether they adopt ML but also in which types of ML applications they prioritize. Such heterogeneity aligns well with the technological context in the TOE framework, suggesting that ML adoption may be influenced by each hospital’s assessment of the relative benefits and costs. Understanding this heterogeneity is essential for designing further policies and implementation strategies that promote equitable and effective ML diffusion across diverse health care settings.

Resource-Based Explanation

Within the TOE framework, the organizational and environmental contexts offer a resource-based explanation for our findings, suggesting that both internal and external resources facilitate or restrict hospital technology adoption. Prior literature has identified such resources as key determinants of EHR adoption [,]. Our findings similarly suggest that hospitals with characteristics commonly associated with greater organizational capacity (eg, larger size, urban location, and system affiliation) were more likely to report ML adoption. Although we did not directly document the decision-making process or measure organizational resources, these characteristics may serve as proxies for financial resources, human capital, IT capabilities, and leadership support. Large, urban, and system-affiliated hospitals typically possess the financial flexibility to purchase new add-on functions from EHR vendors or develop custom ML algorithms tailored to their specific needs. Their established IT infrastructure facilitates efficient ML adoption within EHRs, and their resource-rich environments often have sufficient staffing and training programs to support implementation. Executive leadership and strategic vision also play a critical role in advancing ML adoption [].

  Further, our findings suggest that the unequal distribution of health care resources acts as a significant barrier to ML adoption for hospitals in rural and underserved areas [-] and that external resource support could facilitate ML adoption in such hospitals. Our subgroup analysis findings showed that health system affiliation and leading EHR contract had stronger effects on small hospitals, partly due to the greater availability of financial and technical support.

Heterogeneous ML Adoption

Although most hospitals tend to adopt both clinical and operational ML functions simultaneously, specific hospitals select particular ML applications in alignment with their technical needs, as the TOE framework suggests. Most notably, for-profit hospitals demonstrated a strong interest in adopting ML regarding scheduling functions, which may be explained by staffing structures and institutional objectives. They more frequently experience staffing shortages [] and higher nursing turnover rates []. In this context, ML-enabled scheduling tools offer a practical solution for maintaining service quality at a lower cost by automating front-office functions and reducing reliance on staff.

These findings suggest that hospitals vary in their perceptions, motivations, and priorities when adopting ML technologies, which are often shaped by their resources and strategic goals. This variability highlights the need for qualitative research and targeted surveys on the motives, facilitators, and barriers to implementing emerging ML technologies among health care providers and health system or hospital managers. For instance, researchers may explore how hospital managers perceive and evaluate specific ML technology, enablers and challenges during implementation, and whether ML models meet their expectations.

Limited Model Evaluation

Our findings highlight a notable gap in current hospital AI practice: limited evaluation of models using local hospital or health system data [-]. Given that ML performance is highly dependent on the distribution and quality of the training data in specific contexts, its real-world effectiveness and validity in dynamic health care environments are often unclear and may degrade for specific patients. Moreover, if the training data contain inherent biases, it may perpetuate or even exacerbate existing systemic disparities among vulnerable populations [-].

Researchers in venues such as the Machine Learning for Healthcare conference and the Conference on Human Factors in Computing Systems have emphasized practical frameworks for evaluating ML quality in care delivery, including real-world effectiveness testing, model actionability, model lifecycle management, and performance tracking across sites and settings [-]. To mitigate risk and support safe integration, developers and scholars recommend tailoring evaluation to the clinical context, engaging clinicians and patients, increasing transparency, labeling bias across patient groups, and fostering a supportive organizational culture [-]. This gap also underscores the need for governmental and industrial initiatives and regulations to promote the safe, effective, and trustworthy use of ML technologies []. Although nascent [-], these efforts could provide critical oversight, standards, and incentives that help minimize risks and encourage responsible implementation.

Policy or Practice Suggestion

Multi-level policy interventions are necessary to ensure equitable ML adoption and bridge the digital divide across health care settings. The federal government may introduce an initiative similar to the Health Information Technology for Economic and Clinical Health Act of 2009, which significantly accelerated the adoption of meaningful EHR use among rural and small hospitals [,]. These interventions could include expanding IT broadband infrastructure in underserved areas or providing direct financial incentives to support ML adoption. Meanwhile, health systems and EHR vendors should consider providing additional financial and technical support specifically for rural and small hospitals. Especially, EHR vendors should prioritize integrating foundational ML functions within the existing EHR systems to reduce the technical barriers of meaningful ML usage. Also, offering discounted or complimentary updates for ML features may be a valuable strategy to support hospitals with limited resources.

Limitation

Our study has several limitations. First, our findings should be interpreted as correlational rather than causal, due to unobserved factors. Second, our analysis relies on secondary data from the AHA, which may be subject to self-report bias. Although our design cannot fully address these biases, the AHA data are widely used in health services research as reliable data sources, and they provide unique and detailed information on hospitals’ adoption of ML in EHR unavailable elsewhere. Future studies could validate the AHA data through primary data collection or cross-validation with objective measures. Third, our measures of ML adoption are self-reported by hospital managers, which may be subject to recall error, social-desirability bias, or heterogeneous interpretations of ML functionality. As a result, hospitals may overreport or misreport their level of ML adoption. In addition, we lack several granular contextual variables such as vendor-specific implementation details, interoperability maturity, IT workforce perspectives, and organizational culture, which provide necessary scenarios to understand ML implementation in practice. To address these gaps, future researchers could leverage qualitative and mixed methods designs (eg, interview or focus groups with clinicians, hospital managers, IT staff, and vendor representatives) to capture nuanced adoption processes, identify facilitators and barriers, and document practical implementation strategies. Fourth, the AHA IT Supplement Survey began collecting ML implementation in 2023, which limited our ability to capture early efforts in ML adoption within EHR. Finally, 44.7% of the AHA hospitals did not complete the IT Supplement Survey, raising concerns about potential nonresponse bias. Although we applied inverse propensity score weighting to balance observed characteristics between respondents and nonrespondents, which could strengthen our findings, the nonresponse bias remains if unobserved characteristics are correlated with both survey response and ML adoption.

Conclusion

In this retrospective study of US hospitals from 2022 to 2024, we observed a high rate of ML adoption within EHR systems and considerable variation across specific ML functions. We also identified several hospital characteristics associated with ML adoption within EHRs, including ownership, hospital size, health system affiliation, CAH status, and metropolitan location. Given that resource availability significantly influences a hospital’s capacity to implement ML technologies within EHRs, our findings highlight the need for policy interventions that provide financial and technical support. Such efforts are essential to ensure that resource-constrained hospitals can adopt emerging health IT innovations and prevent the widening of digital and health disparities across the health care system.

To ensure accuracy, fairness, and safety, health care administrators and policymakers must prioritize not only the adoption of ML technologies but also the ongoing monitoring and evaluation of these tools. Equitable access to these technologies must be a key focus, with targeted support for hospitals facing barriers to adoption. By fostering inclusive and transparent approaches to ML adoption, we can maximize its potential to improve care delivery, reduce disparities, and enhance health care outcomes for all patients.

Foley Hoag served as legal counsel to Connectbase, a leading provider of data and software to the communications infrastructure market, in a strategic growth investment in the company from Equality Asset Management (EAM), a Boston-based private equity firm. EAM’s investment will enable Connectbase to accelerate its product innovation and expansion.

The firm also advised Connectbase in the acquisition of Cloud Age, an industry-leading telecom invoice automation provider, which will expand Connectbase’s service offerings. EAM’s investment in Connectbase supported the Cloud Age acquisition.

Corey Brown and Shane Safer led the Foley Hoag team advising Connectbase on the strategic investment from EAM, with support from Erica Rice, Barbara Klepper, Earl Mellott, Silvia Mavares, Sophia Al-Khamis and Alexandra Levay. 

Erica Rice led the Foley Hoag team advising Connectbase in connection with the acquisition of Cloud Age, with support from Sarah Nyaeme, Earl Mellott, Alexandra Levay, Sophia Al-Khamis and Paul Tetenbaum.

Learn more about the transactions here.