Stay ahead of the headlines

Feb 8, 2024·Gizmodo

Google has released its next-generation AI model, Gemini, integrating it into Google Assistant, launching a Gemini mobile app, and offering a premium AI subscription. Google renamed its chatbot from Bard to Gemini, making it available for voice interaction. Gemini Ultra, the company's most powerful AI, costs $19.99/month and outperforms human experts on MMLU, a standardized test covering multiple subjects. The Gemini rollout gives Google users a taste of the AI-centric future of computing.

Feb 8, 2024·Forbes

AI is transforming private equity by enhancing data analysis, deal sourcing, risk assessment, due diligence, and portfolio optimization. Major firms like KKR & Co. and Blackstone use AI to analyze market trends and make more informed decisions. Yet, AI integration poses challenges around data privacy, ethical use, and need for regulatory frameworks. Future vision includes AI identifying potential investments and predicting global economic shifts. AI integration is a leap into a future of unparalleled potential and requires embracing innovation, efficiency and growth.

Feb 8, 2024·Medical Futurist

AI's integration into healthcare raises questions about who pays for its deployment and use. Some cases see patients offered AI-enhanced care at additional cost. The economics of AI versus traditional medical practices remains complex, with cost-effectiveness depending on specific scenarios and services. AI's integration may create disparities in healthcare access, with wealthier persons, countries, and even languages having more access to AI-enhanced services. Despite these disparities, less developed countries could leverage AI to bypass expensive infrastructures, providing a possible answer to healthcare personnel shortages. Healthcare worldwide is grappling with AI's integration, with regulatory measures aiming to balance innovation, accessibility, and fairness.

Feb 7, 2024·LinkedIn

Bipartisan Congressional Digital Health Caucus launches to inform and engage policymakers on #digitalhealth. Discussion included AI's impact on healthcare, like natural language processing improving machine learning accessibility, AI potential in research and improving administrative inefficiencies, and necessary public-private partnerships. AI seen as potential staff gap filler, not healthcare professional replacement.

Feb 7, 2024·AMA

The AMA Update discusses healthcare topics including COVID-19, medical education, advocacy and burnout. Guest Margaret Lozovatsky, AMA's new vice president of Digital Health Innovations, talks about AI's potential to improve healthcare through telehealth and enhancing medical office workflow. She emphasizes the importance of involving physicians in the tool development process to create an asset rather than a burden. The discussion also covers how AI may impact the patient-doctor relationship, stressing augmented rather than artificial intelligence.

Feb 6, 2024·CNN

Finance worker tricked into paying $25m to deepfake posing as company CFO on video call. Despite initial phishing suspicions, authenticity convinced worker to approve transaction. Deepfake technology concerns rising globally.

Feb 5, 2024·CXOToday

AI integration in healthcare revolutionizes disease identification and treatment, improving patient care and fostering sector collaboration. AI assists recovering hospitals, aids during COVID-19, and supports underserved areas with portable X-ray machines and diagnostics, particularly for TB. It also aids remote areas with minimal resources through image interpretation and diagnostics, helping clinicians who may lack disease-specific training. AI has also significantly contributed to TB elimination, providing healthcare professional support, expediting diagnostics and treatment, and reducing patient visit frequency. Widespread AI adoption may help achieve TB elimination by 2025.

Feb 5, 2024·Forbes

Generative AI is revolutionizing healthcare and life sciences, making vital data-driven advancements. Examples include speeding up cancer drug design and accelerating research with large language models. Beyond traditional machine learning and analytics, generative AI absorbs vast data, aids clinicians and researchers, and can drive our health at personal and communal levels. Applications include improving clinical documentation, enabling early disease diagnosis, personalizing treatment, assisting drug research, accelerating clinical trials, and fostering innovation. The challenges lie in data protection, data hygiene, cost maintenance, and customization of AI applications. Generative AI's impact will eventually be felt at the human level as bodies and lives benefit from the enabled advances.

Feb 5, 2024·HealthLeaders Media

Healthcare firms setting AI strategy review committee need careful planning, incorporating skeptics for comprehensive view of tech's use. HealthLeaders AI NOW panel suggests composition: 7-12 members from CFO's team, legal and compliance, medical ethicist, behavioral science, and clinicians. Advise occasional contrarian to challenge group think and promote innovative ideas. Open communication and transparency essential among team. Rapid AI growth requires vigilant monitoring to prevent misuse, potential errors, and ensure effective tech application. Good governance means evaluating vendor's AI tool effectiveness and avoiding duplicate solutions. Health systems should anticipate tech mistakes, apply guiding principles, and engage all affected parties when errors occur, for system improvement. Advocate balanced approach to adopting AI, eyeing its potential in healthcare improvement and life-saving.

Feb 4, 2024·Yahoo News

"Microsoft Q4 profit rocketed 33% due to AI, cloud-computing investments, primarily in cloud-computing unit. Net income was $21.87 billion, exceeding expected $2.79/share. Revenue surged 18% to $62.02 billion, surpassing predictions. Microsoft’s expected ad revenue growth is 12%, potentially adding $14.93 billion in worldwide ad revenues. The results include Activision Blizzard's finances, acquired for $69 billion. The merger increased revenue growth by four points but trimmed operating profits by $440 million. Despite initial share drop, analysts believe investors are worried about Microsoft's aggressive investment plans."

Feb 3, 2024·Medscape

Emergency medicine doctors propose everyone having personal emergency doctors, reducing over and under-triage at hospitals. Current systems don't support this due to insurance companies' structures. Telemedicine growth, integrated systems like Kaiser Permanente and AI-powered kiosks offer better access but have limitations. Concierge medicine provides personal care and accessibility but comes with substantial cost. Innovations reducing costs, fulfilling personal care needs are imperative in healthcare's future.

Feb 2, 2024·DarkReading

JCI spent $27M remediating a September 2023 ransomware attack. The incident involved data exfiltration and locked internal IT infrastructure. Despite concerns, JCI's smart-factory installations remained unaffected. Investigation and remediation are ongoing.

Feb 1, 2024·AAMC

AI tools, like DAX, record doctor-patient conversations and generate written notes, freeing doctors from note-taking during visits. This technology potentially improves the quality of care, reduces documentation time, and alleviates physician burnout. Limitations include transcription errors, omission of crucial medical details, and potential misuse of recorded conversations. Reported benefits include more efficient consultations, accurate notes, and reduced time spent on documentation. Some physicians express concerns about the completeness of the notes and patient privacy. The use and development of such tools, partnered with various EHR platforms, continue.

Feb 1, 2024·Deloitte

Healthcare executives incorporating generative AI into workflows may overlook key success factors if focusing solely on data. Survey reveals 70% focus on data availability, quality, compliance, security, and privacy but neglect important considerations such as effective governance, building consumer trust, and workforce upskilling. Successful integration of AI requires a broader strategic approach.

Feb 1, 2024·William Fry

AI Act's final text out; prohibits AI systems distorting behaviour, exploiting vulnerabilities, inferring biometric categories, evaluating/classifying individuals unfairly, creating facial recognition databases untargetedly and assessing criminal risk based solely on profiling. High-risk AI systems need efficient risk management, documentation, registration, appropriate mitigation measures for feedback loops, and strong human oversight. Act applies to EU-based deployers, AI system outputs used in EU, and general-purpose AI models. Non-compliance warrants heavy administrative fines. Businesses should evaluate AI systems for compliance, document meticulously, implement disclosure measures, invest in human oversight, and communicate transparently regarding AI deployment in workplaces.

Feb 1, 2024·AMA

40% of doctors consider leaving their workplaces due to burnout, low value perception, and heavy clerical burdens. To retain physicians, health systems need to reduce burnout, value doctors, decrease clerical tasks, and smartly use technology like AI to structure tasks enabling doctors to focus on patient care. BindingFlags: "Dr Tina Shah, Chief Clinical Officer at Abridge, suggests that AI can handle repeated, low-level tasks easing workloads."

Jan 30, 2024·JAMA Network

Generative AI, specifically Large Language Models (LLMs), are advancing in the medical sector, aiding in summarizing patient data. However, without FDA oversight, they could reach clinics without safety and efficacy checks. Current EHRs' inefficient info access and excessive content lead to physician burnout and clinical errors, which LLMs could reduce. Variation in LLM-generated summaries might impact clinician decisions. These AI tools require comprehensive standards testing and more explicit FDA regulation given their capacity to alter clinical interpretations and promote potential errors. Existing FDA regulatory safeguards don't clearly cover the unique risks of language-based AI. Transparent standards development and clinical studies are crucial for safe LLM application in clinics.

Jan 29, 2024·CIO

CIO priorities of 2024: data control, balancing innovation and operational excellence, cybersecurity, talent development, AI deployment, increasing digital acumen, addressing AI talent shortage, data governance, value creation, and cost optimization. Emphasis on talent, AI safety, achieving balance, and strategies to better inform and protect business.

Jan 29, 2024·News Medical

Study demonstrates large language models (LLMs) like GPT-3.5 and GPT-4 can simulate clinical reasoning with prompt engineering, improving integration into healthcare. Comparing traditional chain-of-thought (CoT) prompting to diagnostic prompts, GPT models use large volumes of text data for tasks like writing clinical notes and medical exams. Tested on revised MedQA and NEJM datasets, GPT-3.5 and GPT-4 exhibit improved reasoning but not accuracy. GPT-4 shows better accuracy overall, 78% with analytical reasoning prompts, Bayesian inferences and differential diagnostic reasoning. These findings help overcome 'black box' limitations of LLMs, improving user trust.

Jan 27, 2024·Krebs on Security

Google struggles with cybercriminals exploiting its search platform with ads distributing malware-laced software applications. Despite a dedicated team and reductions in these threats, bad ads linked to malware persist. Search results often feature malicious ads, tricking users into downloading harmful software versions. The cybercriminals constantly evolve methods to evade Google's anti-abuse radar. Some malicious domains imitate popular software titles to deceive users. The malicious download sites regularly alternate between serving benign and harmful software. Google has removed violating ads and will continue to enforce its protections. Site-blocking by Google's Safebrowsing technology exists but has not been fully applied. Google removed 5.2 billion ads and suspended over 6.7 million accounts in 2022.