Stay ahead of the headlines

Apr 10, 2024·JAMA Internal Medicine

This study, published in JAMA Internal Medicine on March 25, 2024, examines the effectiveness of an AI-enabled intervention aimed at detecting clinical deterioration among hospitalized patients to reduce escalations in care, such as rapid response team activation, ICU transfers, or cardiopulmonary arrests. Conducting a cohort study of 9,938 patients at an academic center using a regression discontinuity design for analysis, the researchers found that exposure to the intervention led to a 10.4 percentage point reduction in the risk of such clinical escalations. This significant finding suggests that integrating AI deterioration models into patient care can decrease the likelihood of emergency interventions, supporting the potential for broader application and further research in different healthcare settings.

Apr 10, 2024·maginative.com

Sam Altman, CEO of OpenAI, and Jony Ive, former Apple design chief, are in the early stages of funding for a new AI device startup aiming to revolutionize the personal AI product market. This venture seeks to secure up to $1 billion in funding, engaging in discussions with major venture capitalists, including Thrive Capital and Emerson Collective. The startup, which has yet to be named, aims to develop a device that goes beyond the conventional smartphone experience, incorporating advanced AI capabilities. While details on the device are scant, it is hinted that the product will not mimic traditional smartphones, potentially aligning with trends in AI-driven hardware such as wearables and voice-enabled assistants. Both Ive and Altman bring significant experiences from their respective fields, promising a product that could redefine daily interactions with artificial intelligence.

Apr 10, 2024·SC Magazine

As email continues to be a primary target for cybercriminals, leveraging it for phishing and business email compromise (BEC) attacks, organizations are increasingly relying on Security Awareness Training (SAT) to educate employees on identifying threats. However, the emergence of generative AI technologies like ChatGPT has enabled even unskilled attackers to create more convincing and personalized phishing emails, posing new challenges for SAT's effectiveness. While SAT remains crucial in combating low-level email attacks, updating these programs to address the advanced capabilities of AI-generated emails and reinforcing them with additional security measures such as multi-factor authentication and email security solutions are essential for providing comprehensive protection against evolving cyber threats.

Apr 10, 2024·Infoworld

Matt Asay contributes to InfoWorld with an insightful piece on the limitations and misconceptions surrounding large language models (LLMs) and their role in achieving artificial general intelligence (AGI). Asay argues that while LLMs have been hyped as precursors to AGI, they fundamentally lack the capacity for true understanding and are confined to tasks involving statistical text analysis. He emphasizes the importance of diversifying AI research beyond LLMs to include other technologies like reinforcement learning or recurrent neural networks, which could offer more promising pathways toward meaningful AI advancements. Asay critiques the current focus on LLMs for dominating the AI landscape, potentially hindering innovation and encouraging a monocultural approach to AI development. He advocates for a broader exploration of AI technologies to foster genuine progress.

Apr 8, 2024·Modern Healthcare

The article discusses the emerging trend of healthcare organizations incorporating Chief AI Officers into their C-suites, highlighting the growing importance of artificial intelligence in the healthcare industry. As AI continues to shape healthcare delivery and management, professionals like Dennis Chornenky at UC Davis Health, Dr. Karandeep Singh at UCSD Health, and Shawn Wang at Elevance Health represent this new and crucial role. Their positions underscore the evolving landscape of healthcare, where AI's integration is increasingly seen as essential for improving patient care, operational efficiency, and strategic innovation within the sector.

Apr 8, 2024·MSSP Alert

In response to cybersecurity concerns, the U.S. House of Representatives has prohibited Congressional staffers from utilizing Microsoft's Copilot for 365 AI chatbot, as determined by the House's Chief Administrative Officer and Office of Cybersecurity due to the risk of data leakage. This decision extends to a previous restriction on ChatGPT use, emphasizing the legislative body's cautious stance on AI tools. Industry professionals from Quorum Cyber and Varonis Systems, however, assert the security and potential of Copilot, highlighting Microsoft’s commitment to data privacy and the development of specialized solutions to ensure Copilot's secure deployment. Microsoft is actively working on enhancing security measures and has launched Copilot for Security to improve user experience with AI tools while ensuring data protection, particularly for federal agencies with stringent security needs. The company plans to introduce more government-oriented tools to address Congress’s security concerns and leverage AI technology for secure, mission-critical applications.

Apr 8, 2024·Financial Times

Google has encountered significant setbacks in the competitive field of generative artificial intelligence (AI), largely due to internal cultural and organizational challenges. These issues have hindered the tech giant's ability to keep pace with its competitors in the AI race, marking a notable struggle for a company that has long been at the forefront of technological innovation.

Apr 8, 2024·Forbes

Surgeons at the Sint-MaartensKliniek in the Netherlands utilized a 3D-printed, patient-specific hip implant from Materialise for a complex case, highlighting a growing trend towards custom medical devices. The use of off-the-shelf implants, which come in generic sizes, often leads to poor outcomes and additional surgeries due to improper fit. The industry has seen a significant increase in demand for custom 3D-printed implants and surgical guides, spurred by advancements in software from companies like Axial3D and Materialise that have reduced the time needed to convert patient scans into accurate 3D models from weeks to a day or less. This shift towards personalization is being driven by the potential for better surgical outcomes, reduced operation times, and quicker recovery periods. Major medical device makers are investing in patient-specific products, with innovations in AI and machine learning streamlining the design process and potentially reducing production times. Despite the advantages, the adoption of patient-specific implants faces challenges, including the need for a robust supply chain and regulatory guidance, with the FDA expected to issue guidelines on in-hospital 3D printing of implants in 2024.

Apr 5, 2024·The Wall Street Journal

Leading business schools, including American University and Wharton, are significantly integrating artificial intelligence (AI) into their curricula, recognizing its essential role in the future of business. With an emphasis on practical application, these institutions are encouraging students to utilize AI to enhance their problem-solving and decision-making skills. The Kogod School of Business is set to introduce AI components in 20 new or reshaped courses, covering a wide range of subjects from forensic accounting to marketing. This shift is in response to the growing demand for AI proficiency among employers, aiming to equip students with the necessary skills to leverage AI in their careers effectively. The incorporation of AI education reflects the broader recognition of its impact across various business operations and the need for future leaders to be adept at using AI tools and technologies responsibly and innovatively.

Apr 5, 2024·Robert Pearl

In "My Journey with AI and Medicine: 'ChatGPT, MD' Is Here," Robert Pearl, M.D., provides a compelling look into the intersection of artificial intelligence and medicine through a detailed preview of his new book. The book highlights a transformative six-month collaboration with ChatGPT, emphasizing the potential of generative AI to revolutionize healthcare. Pearl recounts the evolution of medical miracles, starting with the pioneering heart transplant in 1967, to set the stage for the future of medicine—a future where AI, specifically generative AI models like ChatGPT, holds the potential to further advance medical practices and improve patient outcomes. Drawing from history, Pearl illustrates how once-impossible medical feats became routine, suggesting the same trajectory for AI in healthcare. The narrative explores the inception and advancements in heart transplantation as a metaphor for the impending revolution in healthcare brought about by AI technologies. By intertwining his personal journey in medicine with the evolution of technological interventions, Pearl makes a compelling case for the integration of AI in medical settings, offering a vision of a future where technology and healthcare professionals work in tandem to redefine the boundaries of possible medical care.

Apr 5, 2024·LinkedIn

The blog post underscores the increasing role of Artificial Intelligence (AI) in healthcare, spotlighting the slow yet growing integration of AI technologies in this sector, ranked 7th out of 12 by a McKinsey AI maturity analysis. It highlights the strategic hiring of Chief AI Officers (CAIOs) within the top health systems in the U.S., despite a current prevalence of only 6%, to spearhead the assessment, strategic application, and integration of AI across organizations. The article segments AI leaders into three phenotypes—operations, transformation, and research and innovation—each with distinct roles, backgrounds, and focal points, suggesting a tailored approach to AI leadership based on an organization's specific needs and goals. It offers reflective questions for healthcare leaders considering AI adoption and emphasizes the significance of dedicated AI leadership in revolutionizing patient care, operational efficiency, and medical outcomes, while also navigating the ethical and governance implications associated with AI deployment in healthcare.

Apr 4, 2024·Becker's Hospital Review

As the healthcare sector increasingly integrates artificial intelligence (AI) into its operations, the debate over the necessity of a Chief AI Officer (CDAO) in hospital and health system C-suites intensifies. Health systems like Richmond, Virginia-based VCU Health argue for the creation of a dedicated CDAO role to oversee data management and ensure the ethical and safe adoption of AI technologies, emphasizing that AI strategy requires focused leadership and expertise. Advocates from institutions such as the Mayo Clinic Arizona stress that a CDAO should be a visionary, navigating the shift towards AI with agility and a patient-centered philosophy. However, some leaders, including prophets from Renton, Washington-based Providence, argue that AI responsibilities should fall within the purview of a well-qualified Chief Information Officer (CIO), treating AI integration as an essential competency rather than necessitating a separate role. This division highlights the varied approaches to governance and strategy implementation as AI becomes increasingly central to healthcare delivery.

Apr 4, 2024·Fierce Healthcare

Epic is set to release an AI validation software suite aimed at enabling healthcare organizations to evaluate and monitor AI models for accuracy, performance, and safety across local patient populations. Designed to automate data collection and provide real-time metrics, this suite seeks to streamline the time-consuming data mapping process, allowing for AI testing and validation at a local level with ongoing monitoring capabilities. This initiative addresses the healthcare industry's challenge in assessing AI technologies, emphasizing the importance of local context in achieving meaningful clinical outcomes. Additionally, Epic plans to make the suite's monitoring template and data schema publicly available to support the deployment of custom and third-party AI models, aiming to facilitate a more transparent and accountable use of AI in healthcare.

Apr 4, 2024·LinkedIn

Bertalan Meskó, MD, PhD, outlines ten predictions for the future of Artificial Intelligence (AI) in healthcare, emphasizing its potential to revolutionize both patient care and the medical profession. The article delves into various aspects of AI's integration into healthcare, including the development of multimodal Large Language Models (LLMs) that mimic clinicians' diagnostic processes, the crucial role of data annotators in building accurate AI models, and the assurance that AI will augment rather than replace physicians. It highlights the shift of AI's role towards taking over repetitive, data-intensive tasks, freeing doctors for more patient-focused responsibilities. Additionally, the piece discusses the importance of understanding and collaborating with AI through anticipation and prompt engineering, addressing challenges such as health equity, regulatory adaptation, and the threat of deepfakes. The article concludes by stressing the need for a combined effort from medical professionals, technologists, and regulatory bodies to navigate the future of AI in healthcare, marking an era where technology enhances human-centric care without replacing the essential elements of empathy and intuition in medicine.

Apr 4, 2024·PYMNTS.com

### AI Reshapes Online Pharmacy With Speed, Accuracy, Convenience The integration of artificial intelligence (AI) into the online pharmacy sector is revolutionizing the way prescriptions are managed, inventories are tracked, and customer interactions are handled, enhancing both speed and accuracy. Companies like Amazon Pharmacy are leveraging AI to offer expedited same-day delivery services in select cities, illustrating AI's potential to improve patient care by shortening delivery times for treatments. AI aids in ensuring the accuracy of prescriptions, boosts productivity, helps in detecting prescription fraud, and optimizes inventory management. Additionally, AI automates routine tasks such as prescription entry and refill reminders, improving operational efficiency and allowing pharmacists to concentrate more on patient care. The technology also plays a pivotal role in predicting medication demand and minimizing stockouts. Looking ahead, AI is expected to offer more personalized healthcare recommendations and assist in disease prevention and medication regimen optimization, promising significant advancements in patient outcomes and healthcare cost reductions. ### 'Swarm' Attacks Challenge Traditional Fraud Defenses The rise of sophisticated AI and machine learning technologies has led to an escalation in payment fraud threats, with fraudsters now capable of orchestrating large-scale social engineering attacks and creating synthetic identities with alarming ease. The democratization of technology allows for these attacks to be both highly credible and scalable, challenging traditional fraud prevention measures. Organizations are finding it increasingly necessary to leverage advanced AI-driven tools and collaborate with specialized fraud prevention companies to combat these threats effectively. Despite the advancements in security measures such as 3D Secure authentication, which can potentially compromise user experience, the battle against fraud requires a delicate balance between stringent security and maintaining optimal customer satisfaction. The article underscores the importance of continuous innovation, data analysis, and consumer vigilance in safeguarding against evolving fraud schemes while emphasizing the instrumental role of technology in fraud detection and prevention strategies.

Apr 4, 2024·Gizmodo

Amazon is discontinuing its "Just Walk Out" technology in Amazon Fresh grocery stores, a system that allowed customers to shop and leave without going through a traditional checkout process. This technology, which employs cameras and sensors to monitor purchases, is being replaced with Amazon Dash Carts and self-checkout counters, primarily due to reliability concerns and the high operational costs associated with the "Just Walk Out" system. Despite its initial portrayal as a fully automated solution, it heavily relied on human reviewers, often leading to delayed receipts for customers. The move to phase out this technology comes alongside layoffs within Amazon's Physical Stores Technology team, indicating a strategic shift towards more cost-effective and reliable customer service technologies. This decision does not affect a small number of Amazon Fresh stores in the UK and some Amazon Go stores where "Just Walk Out" technology will continue to be used.

Apr 3, 2024·Forbes

Michael Dowling, CEO of Northwell Health, disputes claims that healthcare systems are slow to adopt technology, advocating for more intentional innovation through a partnership with Aegis Ventures. Aegis, alongside a consortium of nine health systems, aims to develop startups that address actual needs within healthcare, facilitating quicker and more effective implementation of new technologies. Their approach allows health systems to have direct equity in startups, encouraging innovations like administrative bots and diagnostics tools. Despite this strategic collaboration, health systems including Northwell maintain flexibility in pursuing separate technological ventures, underscoring the complementary rather than exclusive nature of their partnership with Aegis.

Apr 3, 2024·OpenAI

OpenAI has introduced a small-scale preview of Voice Engine, a model capable of generating natural-sounding speech using text input and a 15-second audio sample, which mimics the original speaker's voice closely. Initially developed in late 2022, the Voice Engine has been integrated into OpenAI's text-to-speech API and ChatGPT Voice, emphasizing the cautious approach towards its broader release due to potential misuse. The article showcases various beneficial applications by trusted partners, including providing reading assistance, translating content in the original speaker's voice for global reach, improving service delivery in remote areas, and supporting non-verbal communication and speech recovery for individuals with disabilities or speech impairments. OpenAI is engaging with partners to ensure responsible use, including consent requirements, and exploring safety measures like watermarking and voice authentication to mitigate risks associated with synthetic voice technology. This reflects OpenAI's commitment to AI safety and responsible AI deployment while highlighting the need for societal resilience against the challenges posed by advanced generative models.

Apr 3, 2024·Health Affairs

Mitchell Tang, Kaylee Wilson, and Ateev Mehrotra's article in Health Affairs Forefront, titled "Paying For AI In Healthcare: Setting The Right Precedent Amidst Growing Use," critically examines the challenges and considerations of incorporating artificial intelligence (AI) into healthcare payment systems. It highlights two AI services, fractional flow reserve computed tomography (FFR_CT) and autonomous diabetic retinopathy screening, which exemplify the complexities of reimbursing AI applications in healthcare. The authors point out the discrepancy between the current fee-for-service payment models, which are primarily cost-based, and the unique cost structures and value propositions of AI services. They argue for treating AI company fees as indirect rather than direct practice expenses to more accurately reflect their cost structure and encourage efficient AI pricing models. Additionally, the piece debates the necessity of compensating providers for the time spent analyzing AI results, suggesting that AI tools should ideally improve the efficiency of diagnosis and information synthesis without requiring separate payment for technology use. The authors caution against setting precedents with current AI billing practices that could stifle innovation or encourage overuse, advocating for a payment strategy that incentivizes the adoption of AI based on efficiency and quality improvements rather than direct reimbursement.

Apr 3, 2024·healthsystemcio.com

John Halamka and Paul Cerrato from the Mayo Clinic Platform address the exaggerated fears surrounding AI, arguing that concerns should focus more on realistic technological challenges rather than dystopian scenarios. They discuss the potential of generative AI tools like large language models (LLMs) to invent content or misuse copyrighted material, highlighting ongoing government and healthcare sector efforts to establish responsible AI standards and Quality Assurance labs through the Coalition for Health AI (CHAI). The authors suggest that the real danger lies not in AI becoming autonomous and overpowering, but in its misuse by unethical developers within the healthcare ecosystem, emphasizing the importance of human oversight and ethical development in AI applications.