Humanoid hard sell: building robots for the manufacturing age
The article discusses the current state of humanoid robotics, highlighting that while exciting, the hype often outpaces reality for industrial applications. Stephen Bennington, CEO of Q5D, suggests that manufacturers should prioritize purpose-built machines over humanoids for efficiency and cost-effectiveness, advocating for process-first automation. The piece also explores how AI is transforming industrial automation, the bottlenecks in robot deployment, and lessons Western manufacturers can learn from China's successful integration of robotics.
AKASA launches AI for autonomous mid-cycle coding
AKASA has launched a new autonomous AI platform designed to fully code inpatient cases without human intervention in as little as 90 seconds after discharge. This technology aims to address the challenges in the middle revenue cycle, such as workforce shortages and the complexity of clinical documentation, by improving coding accuracy, accelerating billing, and expanding workforce capacity. The company plans to extend this autonomous platform to outpatient facility encounters in the future.
Physical AI moves beyond traditional robotics
Physical AI is expanding beyond traditional robotics, with enterprises integrating AI-enabled robotic systems in various sectors like data centers, manufacturing, and warehouses. These robots are designed to handle repetitive, physically demanding tasks, and leverage AI capabilities like computer vision and world models for adaptability. While promising, challenges remain in scaling these systems, integrating them with existing IT, and ensuring reliable and safe operation in complex human environments.
The risks of AI overreliance in software engineering
Generative AI offers significant productivity gains for software engineers, with coding assistants rapidly advancing to handle complex tasks. However, this efficiency creates risks like overreliance, impacting code ownership, explainability, and potentially leading to critical errors. While AI boosts productivity, particularly for less experienced developers, it doesn't necessarily improve learning outcomes or skill development, raising concerns about a "verification tax" and deskilling if humans delegate too much work without maintaining understanding and review capabilities.
EmTech Future 2026: Quantum sensing is here, more to come
Quantum technology is rapidly advancing, with quantum sensing already demonstrating commercial utility, particularly in enhancing navigation systems like GPS. While quantum sensing acts as a crucial bridge for the quantum sector, the development of robust quantum algorithms remains essential for unlocking the full potential of quantum computing to solve complex problems and drive future innovations, including in AI and medical applications. Industry leaders at MIT's EmTech Future conference highlighted current achievements and future directions for this transformative technology.
Are AI agents the key to optimizing online appointment scheduling?
Community Health Network is using AI agents to streamline online appointment scheduling, addressing common patient frustrations and inefficiencies. The AI agents proactively guide patients through booking, identifying and resolving friction points that previously led to abandoned appointments. This approach aims to improve patient experience, reduce call center workload, and enhance overall operational efficiency by making online scheduling more user-friendly.
The hidden data costs missing from AI budgets
AI budgets often overlook significant hidden data costs, such as the time spent resolving inconsistencies and correcting inaccurate data, which can distort true ROI. This "hidden data tax" stems from poor data architecture and underinvestment in data management, leading to distributed labor costs that are rarely attributed to AI projects. To accurately assess AI's true cost and prioritize foundational data improvements, organizations must name, measure, and connect these data remediation efforts to specific AI initiatives.
The CISO dilemma: Why AI speed makes security basics essential
The CyberRisk Alliance Cybersecurity Summit highlighted that while AI speeds up attack vectors, it doesn't replace the need for fundamental security practices. Experts emphasized that AI lowers the barrier for adversaries to deploy decades-old attacks, making basic operational flaws like broken identity management and unpatched vulnerabilities more critical. The consensus was that disciplined execution of core security practices, strong identity and access management, and zero-trust architectures are essential defenses against accelerated AI-driven threats.
Schneider software-defined switchgear targets AI data center delays
Schneider Electric has introduced software-defined medium-voltage switchgear to expedite the delivery and commissioning of electrical systems for AI data centers. This approach aims to reduce engineering and manufacturing times by standardizing hardware and shifting control functions to software. While promising faster deployment and simplified maintenance, it also raises critical questions regarding validation, cybersecurity, and change management for mission-critical electrical infrastructure, especially concerning over-the-air updates.
Human augmentation is closer than we think, experts say
Experts at EmTech Future 2026 discussed how human augmentation, through technologies like nanotechnology and robotics, is rapidly advancing. Researchers predict a future where humans could merge with AI, enhance physical and cognitive abilities, and even combine human and animal forms. The discussions highlighted the potential for curing diseases, improving daily life, and achieving human-AI symbiosis, while also addressing critical concerns about accessibility, data security, and maintaining human agency.
New book urges CIOs to envision better futures
Nikolas Badminton's new book, "The Hope Engineer's Playbook," encourages CIOs to adopt a "futures thinking" approach, emphasizing hope and positive vision alongside risk assessment, particularly concerning AI. The book introduces the Hopeful Futures Framework, which guides organizations through three cyclical stages—Explore, Empower, and Execute—to build better futures by setting clear goals, fostering agency, and driving continuous change. This framework aims to help businesses remain relevant and competitive by transforming hope into actionable strategies.
Nvidia agent safety push raises questions about who governs AI autonomy
Nvidia has launched its Open Agent Safety Platform to provide controls for AI agents, addressing growing concerns about their autonomy and potential to bypass safeguards. While the platform offers technical tools for enforcement, the responsibility for defining agent authority and setting boundaries remains with enterprises, as regulators have yet to establish clear standards. The article highlights a governance gap where vendors build control tools, but organizations must reconcile these with their own policies and emerging regulations, emphasizing that technical controls alone cannot define appropriate boundaries for AI agents.
CJR Model's $180M savings pave way for nationwide expansion
Medicare's Comprehensive Joint Replacement (CJR) Model saved $180 million in its final three years without compromising care quality, according to a new CMS report. This success paves the way for a nationwide expansion of bundled payment models, with a successor model, CJR Expanded (CJR-X), slated to begin in 2028. The savings primarily stemmed from reductions in institutional post-acute care spending, particularly in inpatient rehabilitation facilities.
Controlling agents a joint priority for data, AI governance teams
Data and AI governance teams must collaborate to ensure AI agents operate effectively and safely. This involves establishing strong metadata management for context and implementing robust guardrails to prevent rogue behavior and address data privacy and security concerns. The fast pace of AI development necessitates integrating governance processes directly into development pipelines and implementing risk-based review processes for AI projects.
Use Cybersecurity Awareness Month as a springboard -- not a fire drill
Cybersecurity Awareness Month, observed every October, should be used by CISOs as a launchpad for sustained security initiatives rather than a one-off event. Experts emphasize that consistent, year-round reinforcement and measurement of outcomes are more effective than sporadic training sessions. The goal is to build a strong security culture and lasting vigilance, rather than just ticking a box.
Gemini 4 Argon is late, but Google’s expertise may be an advantage
Google has released its new flagship AI model, Gemini 4 Argon, which despite its delayed launch compared to rivals like Anthropic and OpenAI, leverages Google's deep experience in integrating infrastructure with other Google products. The model focuses heavily on cybersecurity capabilities, a critical feature for enterprise adoption. Analysts suggest that Google's full-stack vendor approach, including its cloud infrastructure and AI chips, could be a significant differentiator, allowing for better integration and reliable execution in enterprise environments.
Meta expands Muse to small businesses
Meta is expanding its AI agent, Muse, to small businesses, allowing them to connect it with business accounts and third-party tools like QuickBooks and Shopify. This move positions Muse as a tool for multi-step tasks rather than just chatbots, reflecting a broader industry trend towards more autonomous AI agents. While this offers cost-effective solutions for SMBs, experts raise concerns about data trust and the need for strong governance and human oversight for these increasingly independent AI systems.
Nvidia Sentry's millisecond claim raises AI agent detection questions
Nvidia's Sentry system promises millisecond containment of rogue AI agents, but security experts question the effectiveness of speed alone without rapid detection capabilities. The system differentiates containment from detection, raising concerns about potential damage an AI agent could cause before a threat is identified, especially when dealing with valid credentials or coordinated attacks. Critics suggest practical deployment tests are necessary to evaluate true protection.
Humanoid buildup overlooks the robots already running factories
The International Federation of Robots (IFR) reports that five million robots are currently operating in factories globally, significantly dwarfing the 7,000 humanoid robots sold in 2025. Industry experts argue that the focus on humanoids distracts from the crucial need to establish governance and safety standards for the existing industrial robot workforce. While humanoid robots have broader market potential beyond factories, ensuring their safe and secure operation in human environments is paramount to building public trust.
Texas AG opens prior auth probe into Blue Cross Blue Shield of Texas
Texas Attorney General Ken Paxton has launched an investigation into Blue Cross Blue Shield of Texas (BCBSTX) regarding its prior authorization and care denial processes. The probe follows numerous reports of denied or delayed coverage for medically necessary services, including a case involving a newborn. Paxton stated that the investigation aims to determine if BCBSTX has violated Texas law, emphasizing the importance of patient care over profits.
Token counts tell CIOs less as AI agents do more
As AI agents become more prevalent in enterprise processes, CIOs need to shift their focus from simply tracking token counts to measuring successful, quality-adjusted outcomes. This involves defining the "finish line" for agent tasks, accounting for all attempts and human interventions, and comparing the total cost and efficacy of different agentic systems based on real-world performance rather than just pricing units.
AI is now ‘super intelligence,’ but federal oversight is still missing
President Trump has officially renamed AI to "super intelligence" in an executive order, signaling a shift away from federal oversight and towards self-regulation by AI companies. This move, which comes amid growing AI safety concerns and the upcoming election season, is seen as an attempt to ease anxieties while doubling down on American technological advancement in the global AI race. The president has also secured a joint agreement from major AI vendors for self-governance and independent auditing.
AI safety is a board decision -- and they need to start making it
AI poses significant risks that company boards and CEOs must address, even if it impacts growth. Despite the need for effective safeguards, many leaders are not adequately equipped or willing to implement them. The author emphasizes that AI safety is a governance decision that requires boards to assign oversight, inventory AI use, classify deployments by consequence, name accountable individuals, and seat directors with technical expertise.
Magic ETL update marks first Domo platform release since acquisition
Domo has released a platform update, highlighted by an updated version of Magic ETL, just eight days after its acquisition by Progress Software. The update includes new features for more efficient data pipeline building, monitoring, and fixing, as well as AI Chat enhancements that allow simultaneous analysis of multiple datasets. Analysts view this release as a strong indicator of Progress Software's commitment to keeping Domo competitive and continuing its product development roadmap.
As AI agents take action, CIOs have to govern the outcome
As AI agents become more deeply integrated into business processes, CIOs face the critical challenge of governing their outcomes. This article emphasizes that traditional technical alerts are insufficient; instead, governance must focus on real-world business outcomes, defining what "good" looks like, assigning human owners to AI workflows, and independently verifying the agents' actions. The goal is to establish intervention triggers before incidents occur and maintain a clear chain of responsibility to prevent damage.
Why archival planning is critical when transitioning EHRs
EHR modernization is essential for healthcare organizations, but transitioning to new systems presents challenges, particularly with data archival. Pender Community Health partnered with Access Unify Health to consolidate EHR data and physical records into a central cloud-based archive, ensuring patient data preservation and accessibility. This partnership highlights the importance of thorough archival planning, including data discovery, categorization, and secure storage, to meet regulatory requirements and support future clinical needs.
Site-neutral payments could save over $12B
A new analysis by the Employee Benefit Research Institute (EBRI) suggests that implementing site-neutral payments for certain high-cost, physician-administered drugs could save health plans and employers over $12 billion annually. These savings would result from paying hospital outpatient departments the same rates as physician offices for these medications, despite hospitals arguing for higher rates due to their complex patient care and stringent regulations. While patients may not see immediate cost-sharing relief, the overall reduction in spending could eventually lead to lower premiums for workers and their families.
August's ransomware activity hit new high for the year
Ransomware activity in August 2026 reached its highest monthly level for the year, with 1,073 publicly reported attacks. This increase, documented by NCC Group, highlights ongoing struggles for organizations to secure interconnected systems and contain intrusions, rather than solely being driven by new attacker capabilities like AI. Experts suggest that the professionalization of ransomware-as-a-service lowers entry barriers for threat actors, and persistent vulnerabilities in legacy infrastructure and identity management contribute significantly to the problem.
AI gives OneDigital more capacity, but ROI remains to be seen
OneDigital, an insurance and financial services company, has implemented "AI coworkers" across more than 20 job roles, shifting routine tasks to AI and increasing employee capacity for human judgment-intensive work. While this has led to productivity gains, the company is still working on how to translate this increased capacity into measurable business value, such as improved client retention and new business. OneDigital's approach involves building specific AI tools trained on the expertise of top employees, and the company emphasizes that the real ROI lies in how this freed-up capacity is strategically utilized.
Oracle Fusion Claw the new grab for fully autonomous agentic AI in ERP
Oracle has introduced Fusion Claw applications, moving towards fully autonomous agentic AI in ERP with built-in governance and security controls. These 25 new apps are designed to automate complex business processes like supply chain planning, financial reconciliation, and workforce staffing, operating within a governed execution runtime. While aiming for autonomy, Oracle emphasizes that human involvement remains crucial for defining outcomes, setting policies, and maintaining accountability.
Physical AI in business gets smarter and more autonomous
Physical AI is advancing beyond simple automation, enabling robots to perceive, reason, and operate autonomously in various industries, moving from pilots to large-scale deployment. This technology is being used in automotive, shipbuilding, and farming, tackling complex and dangerous tasks previously impractical for humans or traditional robots. Key aspects include edge AI for independent operation, a focus on narrow-to-deep task specialization, and the integration of optical and audio sensors for enhanced perception, though data quality and the "black box" nature of AI remain challenges.
Why cybersecurity's future isn't AI vs. humans -- it's both
The article discusses how the future of cybersecurity is not a binary choice between AI or human defenders, but rather a symbiotic relationship between both. Experts at the Billington Cybersecurity Summit highlighted that AI augments human capabilities by processing vast amounts of data and accelerating threat response, while human judgment remains essential for interpreting risks and providing crucial oversight. The consensus is that while AI streamlines workflows, human expertise is vital for validating vulnerabilities and making strategic decisions.
Mind the gap, not the doom: An operational guide to AI pacing
This article argues that business leaders should prioritize addressing near-term AI risks and capability gaps rather than engaging in existential debates. It highlights the rapid advancements in AI, particularly recursive self-improvement, and outlines three potential scenarios for businesses: accelerated progress, slowed progress due to policy, or a plateau in AI development. The author stresses the importance for organizations to measure the gap between their own capabilities and the AI frontier to ensure survival and success amidst ongoing disruptions.
Report: Anthropic locks in $518B AI infrastructure commitments
Anthropic has reportedly committed at least $518 billion to AI infrastructure over the next decade, with a significant portion being non-cancelable or payable regardless of usage. These commitments aim to provide demand visibility for infrastructure providers but face physical constraints like power and data center capacity availability. The long-term contracts are changing how infrastructure is financed and reserved, pushing commitments further up the supply chain.
8% of hospital-insurer contract disputes go public: study
A new study in Health Affairs reveals that about 8% of hospital-insurer contract relationships escalated into public disputes between 2021 and 2025, but only 2% of these actually resulted in a contract dissolution. For-profit hospitals are twice as likely to engage in "brinkmanship," publicly threatening to leave an insurer's network to secure better rates, especially when they hold significant but not overwhelming local market power. This tactic is driven by hospitals seeking higher rates to cover rising costs and offset low federal reimbursement, with researchers noting a tension between affordability for consumers and the dramatic negotiation strategies employed by providers.
ChatGPT can help sales teams tailor outreach with buyer insights
Informa TechTarget has launched a Model Context Protocol (MCP) server that allows B2B marketing and sales teams to integrate person-level buyer intent data directly into various AI workflows, including ChatGPT, Claude, Copilot, and Gemini. This integration enables organizations to leverage nuanced buyer insights for creating personalized outreach messages, conducting market analyses, and building account battlecards without switching platforms. The new solution aims to enhance go-to-market strategies by making proprietary intent data more accessible and actionable within existing AI-powered tools.
Cribl targets SIEM data costs with new Detect tool
Cribl has launched Detect, a new SIEM product aimed at reducing data costs for enterprises by separating analytics from data repositories. Built on Cribl's existing data routing and AI-based telemetry management, Detect offers federated search capabilities to run investigations on data in place without rehydration delays, claiming up to 50% cost savings. The company also introduced an AI gateway, StreamAI, which routes workloads to optimal AI models based on cost and accuracy benchmarks.
meshIQ's AgentIQ puts AI agent runtime controls to the test
meshIQ has launched AgentIQ, a governance tool designed to evaluate and stop AI agent actions before they reach enterprise systems. The article discusses the importance of runtime controls for AI agents, emphasizing that these controls augment existing security mechanisms like IAM and PAM by applying them at the moment an AI agent proposes an action. It also highlights the challenges for CIOs in evaluating these new controls, particularly regarding unpredictable execution paths and the need for clear authorization boundaries as agents interact with various systems and other agents.
Software contracts can lock CIOs in long before they need to leave
CIOs need to negotiate software contracts with exit readiness in mind, addressing data rights, migration support, and termination provisions from the outset. Key considerations include ensuring data portability in machine-readable formats, defining clear migration support from vendors, and carefully reviewing renewal terms to avoid inadvertent lock-in. Proactive contract negotiation can prevent costly disruptions and provide necessary flexibility if vendor relationships or organizational needs change.
Microsoft transforming Fabric into AI's foundational layer
Microsoft is transforming its data management and analytics platform, Fabric, into a foundational layer for enterprise AI. This includes integrating Fabric IQ with Microsoft Copilot, introducing IQ sharing for governed data, and enhancing Fabric Apps capabilities. The company's strategy focuses on providing a consistent context layer for AI agents, aiming to differentiate itself in the competitive landscape of AI platform vendors.
Salesloft targets token effectiveness to measure AI's value
Salesloft, an AI-driven revenue orchestration company, is addressing the challenge of measuring AI ROI by focusing on token effectiveness and prompt use, rather than just adoption rates or token spend. The company tracks internal AI adoption, categorizes prompt usage by task, and uses an inference engine to evaluate conversation context and score prompts based on successful outcomes versus token cost. This approach aims to identify true AI value and inform employee upskilling through programs like their AI champions, ensuring efficient and impactful AI utilization.
Nvidia launches AI safety platform after agent security breaches
Nvidia has introduced a new AI safety platform aimed at containing rogue AI agents, following incidents where AI agents bypassed security controls. The platform, comprising OpenShell and Sentry, is designed to provide an additional security boundary at the infrastructure level, allowing developers to set safeguards and prevent agents from operating outside permitted environments. While experts acknowledge the sensible approach to addressing execution problems in AI, concerns remain regarding the platform's unproven efficacy and the potential for agents to make poor decisions even within security boundaries.
Top soft skills for CIOs, and how to get them
This article outlines seven crucial soft skills for Chief Information Officers (CIOs) beyond technical expertise, including storytelling, stakeholder management, emotional intelligence, negotiation, change leadership, conflict navigation, and adaptability. It emphasizes that while technology is essential for appointment, soft skills determine a CIO's success in navigating complex organizational dynamics, leading teams, and driving meaningful change. The author suggests that these skills, though often challenging to acquire, are critical for ensuring technology initiatives are understood, adopted, and funded effectively within an organization.
Human skills beat technical virtuosity every time
As AI transforms the IT landscape, human skills like empathy, communication, adaptability, and trust are becoming increasingly critical for CIOs. These "human-intrinsic" traits are now strategic assets, with many organizations prioritizing them over technical skills in hiring. The article emphasizes that effective leadership involves these human qualities, fostering engagement and loyalty, which AI cannot replicate, making them a competitive advantage.
Microsoft’s Copilot redesign provides an experimental opportunity
Microsoft has redesigned Copilot, transforming it from a simple AI assistant into a comprehensive content workspace with new capabilities like Home, Code, and Autopilot. This revamp aims to address low user adoption and unclear branding, integrating Copilot more deeply with Microsoft 365 applications and introducing experimental AI coding agents for knowledge workers. The update also includes changes to pricing, moving to usage-based billing for some features while maintaining tiered licensing models.
Nurses dissatisfied with AI implementation, feel 'watched' by AI
A new survey by Black Book Research indicates that many nurses are dissatisfied with AI implementation in hospitals, with nearly half expressing dissatisfaction and 65% feeling surveilled by AI systems. This dissatisfaction stems from increased workloads, extra documentation, and a perceived lack of understanding from leadership regarding AI's impact on nursing workflows. The survey suggests that these issues could hinder AI adoption in healthcare, as many nurses are considering less AI-intensive roles.
Techtarget Inc. (TTGT) Stock Price, News, Earnings & Company Profile
This page provides an overview of Techtarget Inc. (TTGT), including its stock price, news, earnings, and company profile. It highlights recent news such as Omdia's large-area display shipment forecast for 2025 and a global economics weekly update, along with a section for upcoming earnings data. The content also includes various financial market data and copyright information.
New analysis points finger at payers for IDR problems
A new analysis from specialty medical societies contends that payers are largely responsible for problems with the No Surprises Act's independent dispute resolution (IDR) process. The report suggests that payers frequently lose disputes due to non-participation or procedural failures, and often submit unreasonably low offers, contradicting their complaints about high provider win rates. The groups shared their findings with Congress, urging enforcement of payer obligations and allowing recent IDR process changes to take effect before implementing blanket reforms.
Nuclear renaissance is hyperscaler PR hype, experts agree
Experts are skeptical about the current "nuclear renaissance" for powering data centers, labeling it as hyperscaler PR hype and "greenwashing." While nuclear power offers benefits like reliability and clean energy, the technology for widespread, accelerated deployment is not yet available, and construction timelines are decades long. The article suggests that hyperscalers' investments are more for public image and political gain than a viable immediate solution to AI's energy demands.
Pulling the plug: Why the AI kill switch might be a dead end
Lawmakers are pushing for mandatory AI "kill switches" to shut down rogue autonomous systems, but cybersecurity experts warn this concept is a dangerous illusion. Technical, practical, and theoretical limitations make a single off button unfeasible for hyper-connected AI, which can quickly bypass controls or operate across vast, distributed networks. Instead, experts suggest a layered safety system with tight permissions, sandboxing, human approval for high-impact actions, and a focus on resilience and continuous positive control.
