A contemporary study conducted by researchers at Harvard University has revealed that advanced artificial intelligence systems are now capable of exceeding human doctors in both diagnosing medical conditions and determining treatment strategies, including in fast-paced and high-stakes emergency room environments. The research specifically accentuates the potential capabilities of modern AI systems in handling complex clinical reasoning tasks that were traditionally considered exclusive to trained physicians.
The findings, published in the peer-reviewed journal Science, are based on a controlled comparison between OpenAI o1 and experienced attending physicians. To ensure realistic testing conditions, the study used 76 actual emergency department cases sourced from Beth Israel Deaconess Medical Center. These cases were evaluated across multiple stages of the diagnostic process, allowing researchers to assess performance under varying levels of available patient information.
At the earliest stage of patient assessment, commonly referred to as initial triage, where clinicians typically have only limited details about a patient’s condition, the AI model demonstrated a notable advantage. It was able to correctly identify either the exact diagnosis or a closely related condition in 67.1 percent of the cases. In comparison, the two physicians involved in the study achieved accuracy rates of 55.3 percent and 50 percent respectively. This suggests that even with minimal data, the AI system was more effective at narrowing down potential diagnoses.
As the diagnostic process progressed and additional clinical information became available during the emergency room evaluation phase, the model’s performance improved further. Its diagnostic accuracy increased to 72.4 percent, reflecting its ability to refine its conclusions with more context. The physicians also showed improvement at this stage, but their accuracy remained lower, at 61.8 percent and 52.6 percent. This stage is particularly important as it mirrors real-world conditions where doctors continuously update their assessments based on new findings.
In the final phase of care, when patients were admitted either to general hospital wards or intensive care units, the AI model continued to outperform its human counterparts. It achieved an accuracy rate of 81.6 percent, compared to 78.9 percent and 69.7 percent for the physicians. Although the performance gap narrowed slightly at this stage, the AI still maintained a measurable edge, indicating consistency across the full diagnostic timeline.
Beyond identifying illnesses, the study also evaluated how effectively the AI system could design clinical management plans. This included decisions such as selecting appropriate medications, including antibiotics, as well as handling complex and sensitive scenarios like end-of-life care planning. Across five evaluated case studies, the AI achieved a median performance score of 89 percent. In contrast, physicians scored significantly lower, averaging 34 percent when relying on traditional clinical resources and 41 percent when supported by GPT-4. This underlines a substantial gap in structured decision-making support.
The researchers acknowledged that while integrating AI into clinical workflows is often viewed as a high-risk approach due to patient safety concerns, its potential benefits are significant. They noted that wider adoption of such systems could help reduce diagnostic errors, minimize treatment delays, and address disparities in access to healthcare services. These factors collectively contribute to both improved patient outcomes and reduced financial strain on healthcare systems.
At the same time, the study emphasizes that current AI systems are not without limitations. Clinical medicine involves more than text-based data. Doctors routinely rely on non-verbal and non-textual cues, such as observing a patient’s physical discomfort, interpreting imaging results, and making judgment calls based on experience. These aspects are not fully captured by existing AI models, which means human expertise remains essential.
The authors further concluded that large language models have now surpassed many traditional benchmarks used to measure clinical reasoning abilities. However, they stress the urgent need for more detailed research, including real-world clinical trials and studies focused on human-AI collaboration, to determine how these systems can be safely and effectively integrated into healthcare settings.
In comments shared with The Guardian, lead researcher Arjun Manrai clarified that the findings should not be interpreted as suggesting that AI will replace doctors. Instead, he described the results as evidence of a major technological shift that is likely to transform the medical field in the coming years.
From a macro industry perspective, this study reflects a developing trend in which AI is increasingly being used to augment clinical decision-making. However, experts continue to caution that challenges such as data bias, accountability, regulatory oversight, and patient trust must be addressed before such systems can be widely deployed. The future of healthcare, therefore, is likely to involve a collaborative model where AI amplifies efficiency and accuracy, while human doctors provide critical judgment, ethical oversight, and patient-centered care.
U.S.-based medical imaging provider SimonMed Imaging has disclosed a cybersecurity incident that compromised the personal data of more than 1.2 million patients earlier this year. The company, which operates nearly 170 diagnostic centers across 11 states, specializes in radiology and imaging services such as MRI, CT scans, X-rays, ultrasounds, and mammography.
Details of the breach
According to information shared with regulators, unauthorized individuals gained access to SimonMed’s internal systems between January 21 and February 5, 2025. The breach came to light on January 27, when one of SimonMed’s third-party vendors reported a security incident that also affected the company. An internal investigation confirmed suspicious network activity the following day.
SimonMed stated that once the attack was detected, the organization acted swiftly to contain the intrusion. Measures included resetting employee passwords, activating multifactor authentication, adding endpoint detection and response (EDR) tools, cutting off third-party vendors’ direct system access, and restricting external network connections to only verified sources. Law enforcement authorities were notified, and cybersecurity specialists were brought in to assist in the investigation and recovery process.
Data possibly exposed
While SimonMed has not disclosed the full scope of data accessed by the attackers, the company confirmed that patients’ full names were among the exposed information. Given the type of data typically stored in radiology systems, the breach may also involve sensitive records such as identification details, medical reports, and financial information.
As of October 10, SimonMed reported finding no evidence that the compromised data has been used for fraud or identity theft. Affected individuals have been offered free identity theft protection services through Experian as a precautionary step.
Ransomware group claims responsibility
Shortly after the breach, the Medusa ransomware group claimed responsibility, listing SimonMed on its leak site on February 7. The group alleged that it had stolen 212 gigabytes of data and released a small sample online as proof. The leaked files reportedly contained ID scans, patient information spreadsheets, billing details, and diagnostic reports.
Medusa demanded a ransom of $1 million, along with an additional $10,000 fee for each day the company delayed payment before full data disclosure. SimonMed’s name has since been removed from the group’s website, which often suggests that negotiations may have taken place. However, the company has not confirmed whether any ransom payment was made.
Growing threat to healthcare organizations
The Medusa ransomware operation, which surfaced in 2023, has been linked to several high-profile attacks on critical infrastructure, including the Minneapolis Public Schools and Toyota Financial Services. In March 2025, the FBI, CISA, and MS-ISAC jointly warned healthcare and education organizations about Medusa’s ongoing targeting campaigns.
Cybersecurity experts emphasize that healthcare institutions remain vulnerable due to the volume of sensitive data they handle. Experts recommend strengthening authentication protocols, monitoring system activity, and maintaining up-to-date security measures to minimize the risk of future incidents.
In September 2024, American Addiction Centers (AAC) experienced a significant cyberattack that exposed the personal and health-related information of 422,424 individuals. The breach involved sensitive data such as Social Security numbers and health insurance details, prompting AAC to take immediate action to address the situation and support those impacted.
The cyberattack occurred over three days, from September 23 to September 26, 2024. AAC identified the breach on September 26 and quickly launched an investigation. The organization engaged third-party cybersecurity experts and notified law enforcement to assess the extent of the attack. By October 3, investigators confirmed that unauthorized individuals had accessed and stolen data during the breach.
Data Compromised
The stolen information included:
AAC assured clients that no treatment information or payment card data was compromised during the incident. While the exposed data could be exploited for identity theft, the company emphasized that there is no evidence linking the breach to fraudulent activity at this time.
Response and Notifications
In December 2024, AAC began notifying affected individuals, with official letters sent out just before the holiday season. These notifications outlined the breach details, the steps AAC had taken to address the incident, and the measures implemented to prevent future occurrences.
To mitigate risks, AAC offered affected individuals complimentary credit monitoring services for 12 months. The organization partnered with Cyberscout, a TransUnion company, to provide identity protection services, including:
Affected individuals are encouraged to enroll in these services by March 31, 2025, to ensure their data remains protected.
Beyond immediate support, AAC implemented enhanced security protocols to strengthen its cybersecurity defenses. The organization collaborated with leading cybersecurity experts to fortify its systems and continues to review and update its measures regularly. Additionally, AAC established a dedicated hotline to assist individuals with inquiries and provide guidance on protective measures.
Proactive Recommendations
Although AAC has found no evidence of identity theft linked to the breach, it urged affected individuals to take the following proactive measures:
AAC’s notification letters include detailed instructions on these steps to help individuals safeguard their personal information against potential threats.
AAC emphasized its dedication to protecting personal information and maintaining transparency with its clients. “We regret that this incident occurred and the concern it may have caused,” the company stated. “We take the confidentiality and security of personal information very seriously and will continue to take steps to prevent a similar incident from occurring in the future.
As investigations into the incident continue, AAC remains focused on strengthening its cybersecurity measures and rebuilding client trust. For further information, individuals can contact AAC’s dedicated hotline at 1-833-833-2770, available Monday through Friday from 8:00 a.m. to 8:00 p.m. Eastern Time.
This incident highlights the importance of robust cybersecurity measures and proactive data protection strategies to safeguard sensitive information in the digital age.
A severe cyberattack on Ascension, one of the largest healthcare systems in the United States, has disrupted patient care significantly. The ransomware attack, which began on May 8, has locked medical providers out of critical systems that coordinate patient care, including electronic health records and medication ordering systems. This disruption has led to alarming lapses in patient safety, as reported by health care professionals across the nation.
Marvin Ruckle, a nurse at Ascension Via Christi St. Joseph in Wichita, Kansas, highlighted the chaos, recounting an incident where he almost administered the wrong dose of a narcotic to a baby due to confusing paperwork. Such errors were unheard of when the hospital’s computer systems were operational. Similarly, Lisa Watson, an ICU nurse at Ascension Via Christi St. Francis, narrowly avoided giving a critically ill patient the wrong medication, emphasising the risks posed by the shift from digital to manual systems.
The attack has forced hospitals to revert to outdated paper methods, creating inefficiencies and increasing the potential for dangerous mistakes. Watson explained that, unlike in the past, current systems for timely communication and order processing have disappeared, exacerbating the risk of errors. Melissa LaRue, another ICU nurse, echoed these concerns, citing a close call with a blood pressure medication dosage error that was fortunately caught in time.
Health care workers at Ascension hospitals in Michigan reported similar issues. A Detroit ER doctor shared a case where a patient received the wrong medication due to paperwork confusion, necessitating emergency intervention. Another nurse recounted a fatal delay in receiving lab results for a patient with low blood sugar. These incidents highlight the dire consequences of prolonged system outages.
Justin Neisser, a travel nurse at an Indiana Ascension hospital, chose to quit, warning of potential delays and errors in patient care. Many nurses and doctors fear that these systemic failures could jeopardise their professional licences, drawing parallels to the high-profile case of RaDonda Vaught, a nurse convicted of criminally negligent homicide for a fatal drug error.
The health sector has become a prime target for ransomware attacks. According to the FBI, health care experienced the highest share of ransomware incidents among 16 critical infrastructure sectors in 2023. Despite this, many hospitals are ill-prepared for prolonged cyberattacks. John Clark, an associate chief pharmacy officer at the University of Michigan, noted that most emergency plans cover only short-term downtimes.
Ascension's response to the attack included restoring access to electronic health records by mid-June, but patient information from the outage period remains temporarily inaccessible. Ascension has asserted that its care teams are trained for such disruptions, though many staff members, like Ruckle, reported receiving no specific training for cyberattacks.
Federal efforts to enhance health care cybersecurity are ongoing. The Department of Health and Human Services (HHS) has encouraged improvements in email security, multifactor authentication, and cybersecurity training. However, these measures are currently voluntary. The Centers for Medicare & Medicaid Services (CMS) are expected to release new cybersecurity requirements, though details remain unclear.
The American Hospital Association (AHA) argues that cybersecurity mandates could divert resources needed to combat attacks. They contend that many data breaches originate from third-party associates rather than hospitals themselves. Nevertheless, experts like Jim Bagian believe that health systems should face consequences for failing to implement basic cybersecurity protections.
The cyberattack on Ascension calls for robust cybersecurity measures in health care. As hospitals consolidate into larger systems, they become more vulnerable to data breaches and ransomware attacks. Health care professionals and patients alike are calling for transparency and improvements to ensure safety and quality care. The situation at Ascension highlights the critical nature of cybersecurity preparedness in protecting patient lives.
Orrick, Herrington & Sutcliffe, the San Francisco-based company revealed last week that that during an attack in March 2023, threat actors stole personal information and critical health data of more than 637,000 data breach victims.
Orrick said that the hackers had taken massive amounts of data from its systems related to security incidents at other organizations, for which he provided legal assistance, in a series of letters notifying those impacted of the data breach.
Orrick informs that the data involved in the breach involved its customers’ data, including those with dental policies with Delta Dental, a major healthcare insurance network that covers millions of Americans' dental needs, and those with vision plans with insurance company EyeMed Vision Care.
The company further added that it had contacted with the U.S. Small Business Administration, the behavioral health giant Beacon Health Options (now Carelon), and the health insurance provider MultiPlan that their data was also exposed in Orrick's data breach.
Apparently, the stolen data includes victims’ names, dates of birth, postal address and email addresses, and government-issued identification numbers, such as Social Security numbers, passport and driver license numbers, and tax identification numbers. Also, information about patient’s medical treatment and diagnosis details, insurance claim like date and service-charges, and healthcare insurance numbers and provider details have been compromised.
Orrick further says that credit or debit card details as well as online account credentials were also involved in the breach.
Since the initial announcement of the breach, the number of affected individuals have been on the rise. In its recent breach notice, Orrick states that it “does not anticipate providing notifications on behalf of additional businesses,” however the company did not specify how it came to this conclusion.
Orrick said in December to a federal court in San Francisco that it reached a preliminary settlement to end four class action lawsuits that claimed Orrick failed to disclose the breach from victims for months after it had occurred.
“We are pleased to reach a settlement well within a year of the incident, which brings this matter to a close, and will continue our ongoing focus on protecting our systems and the information of our clients and our firm,” added Orrick’s spokesperson.