Category: What’s New in Technology

Insurance for AI Risk: Is It Time to Consider AI Liability Coverage?

Over the past few years, artificial intelligence (AI) has evolved from a futuristic concept into a core engine of modern enterprise strategy. Organizations across every major industry are now using AI to automate complex workflows…

4 min read

Insurance for AI RiskOver the past few years, artificial intelligence (AI) has evolved from a futuristic concept into a core engine of modern enterprise strategy. Organizations across every major industry are now using AI to automate complex workflows, augment customer service operations, drive predictive decision-making, and unlock greater operational productivity.

Understanding AI Risk

AI is not an easily defined category, as it spans several dimensions that traditional risk frames are not built to accommodate. The Gallagher report, Smart Systems, Blind Spots: Rethinking Insurance for the AI Era, found that the pace of AI adoption surpassed the insurance industry’s capacity to develop responsive products.

What makes AI unique is that risks associated with it emerge from the way systems learn, generate outputs, and make decisions to influence customers, employees, and business outcomes.

Modern businesses face several distinct risk vectors:

  • Biased or discriminatory decisions
    Automated recruitment, lending, or credit-scoring models trained on flawed data can produce systematically unfair outcomes. This can result in regulatory penalties, civil rights litigation, and damaged brand reputation.
  • Hallucinations and inaccurate outputs
    AI models can confidently generate inaccurate or misleading information. A customer-facing AI assistant that provides incorrect financial, legal or medical guidance could create significant liability exposure.
  • Intellectual property and copyright disputes
    Models trained on vast, unvetted datasets reproduce copyrighted material, exposing organizations to costly intellectual property infringement claims.
  • Data privacy violations
    Unintentional exposure of proprietary trade secrets or personally identifiable information (PII) during model training can trigger regulatory investigations under frameworks such as the EU AI Act, the General Data Protection Regulation (GDPR), or state-level privacy laws.
  • Cybersecurity vulnerabilities
    AI introduces new attack vectors, including prompt injection, data poisoning, and model extraction. Malicious actors can exploit these to compromise business integrity.
  • Financial losses
    Autonomous trading agents or algorithmic pricing models operating at high speeds can execute erroneous transactions, leading to immediate financial losses.

Why Traditional Insurance May Not Be Enough

Existing coverage was not designed for current AI issues. Cyber policies were designed around data breaches and network intrusion. This does not cover an AI model making a biased hiring decision or fabricating a financial projection.

Professional indemnity and E&O policies assume a human professional exercised judgment. So, when an algorithm makes a mistake, an insurer may dispute whether the policy was intended to respond. For general liability policies, the focus is on bodily injury and property damage. If an AI program causes bodily injury, insurers can debate whether the policy applies.

Several incidents have caused some insurance companies to exclude AI from their corporate policies. For instance, Google was sued by a Minnesota-based company after its AI Overviews feature named it as a defendant in a lawsuit. This is just one case that highlights the growing concern around “silent insurance” when policies do not explicitly address AI-related risks. However, businesses may assume they are covered when they are not.

The challenge is compounded by the rapidly evolving legal landscape, with governments worldwide introducing new regulations.

The Rise of AI Liability Coverage

In response, a new category is beginning to take shape. This is AI liability insurance. These policies are designed to explicitly address the development, deployment, and use of AI systems. While offerings may vary across providers, AI liability covers incidents such as AI-driven discrimination claims, IP infringement from generative outputs, financial losses from automated decision-making, and regulatory penalties tied to AI non-compliance.

Insurers are approaching underwriting as they did with early cyber policies. They are starting cautiously, requiring detailed disclosure of how AI is used, existing governance controls, and how models are tested and monitored.

Beyond Insurance: Building Comprehensive AI Resilience

Insurance alone cannot eliminate AI risk and should not be a substitute for operational resilience. Organizations building genuine AI resilience are investing in:

  • Formal AI governance frameworks
  • Meaningful oversight of consequential decisions
  • Ongoing model monitoring and auditing
  • Employee training on responsible AI use
  • Clearly articulated responsible AI principles
  • Tested incident response plans specifically for AI-related failures.

A well-governed AI program will also make a business significantly more insurable, as underwriters increasingly price risk based on demonstrated controls.

Conclusion

AI has become one of the greatest sources of competitive advantage as well as a new source of liability. As regulatory scrutiny increases and AI-driven decisions become more consequential, executives must broaden their understanding of enterprise risk. Insurance should not be viewed as a substitute for governance, oversight or responsible AI practices.

For businesses increasingly relying on AI, the question is no longer whether AI creates liability risk, but whether existing insurance is equipped to respond to it. 

The Death of the App: Why Your Business Will Sideline SaaS Dashboards

For two decades, enterprise software has been built around a simple assumption: people log into multiple applications to retrieve information, make decisions and complete work. A CRM, a project tracker, a business intelligence…

4 min read

Sideline SaaS DashboardsFor two decades, enterprise software has been built around a simple assumption: people log into multiple applications to retrieve information, make decisions, and complete work. A CRM, a project tracker, a business intelligence dashboard, a support ticketing system, and more. All this is because these applications operate in isolation.

There is a shift whose intention is not eliminating SaaS applications. It’s about eliminating the need to constantly switch between them.

Why Dashboards Existed

Dashboards were built because software couldn’t interpret business intent. Humans had to retrieve, interpret charts, and decide what to do next. While dashboards were designed for human navigation, these static SaaS front ends are being replaced by dynamic, real-time interface synthesis.

The dashboard model worked when companies relied on a handful of applications. Today, enterprises manage hundreds of SaaS tools. An average large enterprise runs multiple SaaS applications – about 291 with large organizations scaling over 400. This makes constant switching a productivity problem rather than convenience.

A Harvard Business Review study revealed that digital workers toggle between different applications and websites about 1,200 times a day. This tool-switching alone costs employees an average of 44 hours per year due to tool fatigue. Meanwhile, most of the enterprise SaaS stack goes completely unused, and this is a weighty business cost.

What is Actually Changing

The shift in business computing is not about adding another dashboard to the stack, but rather usurping its purpose. The enterprise interface is beginning to shift toward intent-native workspaces, reducing the need to navigate traditional dashboards for routine work.

In comes agentic AI, which collapses the decision chain. Instead of opening a chart to figure out what it means, the user states an intent and an agent queries the underlying systems directly, synthesizes across them, and gives the user an answer or takes the action itself. For example, instead of a user logging into five different systems, a finance agent pulls real-time vendor invoices from an ERP, a legal agent scans contract terms, and a risk agent cross-references historical delivery delays. All coordinated by an orchestration layer.

Generative user interface (GenUI) technology pairs with this orchestration. Instead of presenting the same dashboard to everyone, a GenUI system generates a temporary interface tailored to the user’s immediate request. Once the task is complete, that interface disappears. If a user inputs their intention, such as checking which supplier poses a risk, the system dynamically renders a clean, interactive panel showing only the relevant vendor risk scores.

A survey by CrewAI on 2026 State of Agentic AI Survey found that adoption of agentic AI is moving fast. Of the 500 senior enterprise executives surveyed, 65 percent are already using AI agents, 81 percent have fully adopted and are actively scaling, and 100 percent plan to expand agentic AI use in 2026.

What Still Matters

Dashboards aren’t disappearing; their role is changing. The shift is not toward a better dashboard; it is to create systems that decide and act directly, with humans overseeing outcomes and not every step. Modern AI-driven operations demand speed that previous tools can’t cope with. Having insights without action is now a bottleneck. Static views, manual interpretation, and the lack of proactive alerts and personalized framing are limitations that drive the shift toward agents.

However, while agentic AI determines what happens next, the dashboards will keep documenting the process. They will also exist mainly as audit trails and compliance records, but not as the primary way work gets done.

What This Means for Your Business

For businesses evaluating software, appearance is becoming less important than accessibility. A polished dashboard matters little if AI agents can’t access its data or trigger actions. As enterprises increasingly rely on AI agents to automate work across multiple systems, software without strong AI integration risks becoming difficult to use, costly to upgrade, and easier to replace.

Logistically, this means businesses should start auditing their software stack for API maturity and AI agent readiness. Before renewing or purchasing new software contracts, a business should evaluate whether the platform has robust APIs, allows AI agents to securely access its data and perform actions, and is built to support an AI-driven workflow.

Conclusion

The biggest disruption is not the end of SaaS dashboards – it’s the end of software that waits for human input. The next generation of enterprise software won’t compete on who has the prettiest dashboard. It will compete on which platform gives AI agents the fastest, safest access to data and actions. Businesses that continue buying interfaces instead of intelligent access may soon find themselves paying for software no one opens.

Beyond Passwords: Why Recent 24B Records Leak is Wake-Up Call for Stronger Authentication

24B Records Leak, Beyond Passwords

4 min read

24B Records Leak, Beyond PasswordsThe recent discovery of a publicly available Elasticsearch cluster, a group of interconnected search servers, containing 24 billion exposed records, is among the largest-scale data breaches, highlighting the troubling reality that passwords have become a weak link in modern digital security.

For years, one of the responses to cyberthreats has been to create stronger passwords, implement password rotation policies, and deploy password managers. Despite all these efforts, credential-related attacks continue to dominate the threat landscape.

The latest threat is a reminder that the problem is not simply password hygiene – but the password itself.

The Weaknesses of Password-Based Security

Passwords were designed for a simpler era of computing. Today, passwords are used to protect everything from corporate networks and cloud applications to banking platforms and healthcare systems. Even with the evolution in computing, the basic principle of passwords remains unchanged. That is, access is granted on a secret that can be stolen, guessed, reused, or shared.

The 24 billion record leak demonstrates the scale of this vulnerability. This means cybercriminals now possess records of usernames, email addresses, login URLs and passwords that can be weaponized against organizations.

The password challenge is made worse by human behavior. Users often reuse passwords across multiple accounts, use predictable combinations, or rely on slight variations of existing credentials. This means a breach affecting one platform can easily become a gateway to many others.

Unfortunately, organizations continue to invest heavily in securing networks, endpoints and applications while still relying on an authentication mechanism that is failing to withstand today’s threat environment.

Why Traditional Defenses Are No Longer Adequate

The greatest danger that arises from a big password leak is credential stuffing attacks. In these attacks, cybercriminals systematically test stolen username and password combinations across thousands of websites and applications using automated tools. Since users frequently reuse credentials, attackers can achieve high success rates with minimal effort. The credential stuffing attacks model allows threat actors to compromise accounts without exploiting software vulnerabilities or bypassing sophisticated security controls.

Even password managers, although valuable, are not the best solution. They help users generate and store stronger credentials, but are not immune to phishing attacks, session hijacking, malware-based credential theft, or social engineering attacks.

Multi-factor authentication (MFA) improves security. However, attackers have increasingly taken advantage of MFA fatigue attacks, SIM-swapping, and real-time phishing proxies.

Simply put, organizations are investing significant resources to protect a flawed authentication model.

Passwordless Authentication: The Next Evolution of Identity Security

The business impact of credential compromise has far-reaching consequences. The solution today is not the use of stronger passwords – but instead, reducing dependence on them altogether.

Passwordless authentication promises more secure methods that are resistant to phishing, credential theft, and reuse attacks. Several technologies are emerging as a replacement for traditional credentials.

  1. Passkeys
    A passkey is a fast identity online (FIDO) authentication credential where, instead of typing a secret word, a user device confirms who they are using built-in security. An example is when you log in to a Google account, and your phone simply asks for your fingerprint or face scan.
  2. Biometric Authentication
    This adds another layer of convenience and security. It includes fingerprint scans, facial recognition, and other biometric identifiers. These allow users to authenticate using characteristics that are unique to them rather than information they must remember.
  3. Hardware Security Keys
    This provides another powerful option. It involves the use of physical devices such as YubiKeys or Google Titan Security Keys that authenticate users through public-key cryptography. Because the private key never leaves the device, it provides strong protection against phishing and credential theft and is widely considered among the most effective defenses against account compromise.

Despite the advantages of these passwordless methods, adoption remains low. Many organizations continue to operate legacy systems designed around traditional username and password models. It is worth noting that the integration of modern authentication frameworks does require significant planning and investment. However, it should be considered as an evolution that requires strategic commitment rather than a quick fix.

Final Thoughts

The recent exposure of 24 billion records is more than another headline-grabbing cybersecurity incident. It is evidence that the password-centric model of digital security is no longer secure. This should prompt organizations still using the traditional password methods to adopt passwordless authentication.

As technology advances, new security challenges will arise, including the emergence of quantum computing and the need for quantum-resistant cryptography. These developments reinforce the lesson that security cannot remain static. The goal is not to predict every future threat, but to build security architectures that evolve with technology. 

Why Human-Centric Marketing Beats AI Spam Every Time

Human Marketing Beats AI Spam

4 min read

Human Marketing Beats AI SpamEvery day, businesses are publishing AI-generated blog posts, automated emails, faceless videos, and social media threads at an unprecedented scale. A prompt can now produce what once required hours of brainstorming and execution.

This probably sounds like every marketer’s dream. However, audiences are becoming more selective, more skeptical, and more emotionally disconnected from brands that sound robotic or overly automated.

This also has made human connection more valuable. The businesses winning today are not those creating the most content, but those creating the content that feels real, personal, trustworthy, and emotionally intelligent. In a world flooded with AI-generated noise, human-centric marketing has become a competitive advantage.

The Internet Has Entered the Era of AI Saturation

AI tools have completely transformed marketing. Businesses now use AI to write ads, generate videos, automate customer support, create product descriptions, analyze customer behavior, and schedule content.

According to recent industry research, generative AI adoption among marketing teams has exploded due to the efficiency and ROI it promises. However, in the pursuit of efficiency, many businesses are optimizing for systems and not people, running the risk of marketing to search engines and not the humans who are typing keywords into them.

Consumers are increasingly noticing that a lot of the content feels emotionally flat, repetitive, or lifeless. As a result, consumers are becoming fatigued with overly polished, mass-produced AI content.

In many ways, AI has lowered the barrier to creating content – while at the same time raising the standard for creating meaningful content. Today, audiences are not looking for more information. They want resonance, relatability, trust, and emotional connection.

The Anatomy of AI Spam: The ‘More is Better’ Illusion

In the early 2020s, the marketing playbook was all about using AI to multiply production, lower cost per asset, and dominate search engines and social feeds through volume.

This resulted in companies adopting generative AI to scale up their monthly content output. But this push for volume has triggered an aggressive case of digital fatigue. Audiences are learning to spot signs of automated copies, such as repetitive sentence structures, unnatural personalization, generic advice, lack of original insight, and emotionally detached messages.

AI spam also suffers from algorithmic sameness as AI models train on identical and widely available data. Unless carefully refined with brand-specific insights and human creativity, many outputs begin sounding identical. This lazy automation has broken consumer trust.

The Hybrid Framework

Choosing human-centric marketing does not mean launching an anti-technology crusade or abandoning modern software tools. The smartest marketing strategies today combine AI efficiency with human intelligence. A marketing team can use generative tools to manage backend activities like sorting data, testing headlines, generating first drafts and automating repetitive workflows.

With the administrative friction taken care of, the marketing team will have time to listen closely to its audience, understand emotional context, tell compelling stories and build authentic trust. Technology in this case will drive distribution scale, while genuine human connection drives sales conversations.

How to Make Your Marketing More Human-Centric

As AI-generated content becomes more common, standing out will require marketers to become more intentional about human connections. Here are practical ways to make marketing feel more authentic and emotionally relevant:

  1. Share real experiences – talk about lessons learned, mistakes made, challenges faced and real outcomes. These experiences cannot be automated convincingly.
  2. Develop audience clarity – understanding not just who your audience is, but what they fear, desire, struggle with and aspire toward.
  3. Human language and storytelling – avoid sounding overly corporate, robotic or excessively optimized. People connect with conversational communication, relatable stories and emotional honesty.
  4. Build around emotion, not features – focus on how customers want to feel, such as confident, respected, productive, safe or inspired.
  5. Show the humans behind the brand – feature team members, founders, behind-the-scenes moments, customer stories and authentic interactions.
  6. Focus on conversations, not broadcasting – respond thoughtfully to comments, engage in discussions and listen to feedback.
  7. Prioritize original thinking – a major weakness with AI-generated content is sameness. Sharing unique opinions, fresh insights and real expertise instead of recycling popular talking points helps a business stand out.
  8. Use AI as an assistant, not a replacement – AI should support human creativity and not replace the human voice entirely. Every outbound message should pass a simple test before it is sent: Would a real, informed person who actually cares about this customer send this message? If the answer is no, the message should not be sent.

Final Thoughts

Human-centric marketing focuses heavily on real customer experience, emotional understanding, and authentic communication. Brands that want to remain successful must continue investing time and energy in understanding real human behavior, emotion, and trust.

 

The ROI of Autonomy: Measuring the Business Value of Agentic AI Workflows

4 min read

Measuring the Business Value of Agentic AI WorkflowsBusinesses are moving beyond basic automation into a new era of intelligent, self-directed systems. While automation helps with streamlining repetitive tasks, agentic AI workflows enable systems to make decisions, take action, and continuously improve with minimal human oversight.

Most businesses adopting agentic AI have no structured way to prove it is working. Although they can feel the difference, they can’t measure it. Without measurement, return on investment (ROI) conversations stall, budgets get cut, and genuinely transformative tools get shelved.

What Makes Agentic AI Workflows Different

Agentic AI workflows are designed to operate with a degree of independence. Unlike traditional automation, which follows predefined rules, agentic systems are goal-oriented.

Once given an objective, they plan, execute, adjust, and complete tasks across multiple steps, tools, and decisions without requiring human intervention. For example, an agentic workflow may pull data from multiple systems, analyze it, draft a report, flag anomalies, and email a summary.

Another example is a supply chain AI agent that not only highlights anomalies but can also reorder stock, renegotiate pricing thresholds, and even reroute logistics as these actions fall within predefined objectives.

Agentic AI can also improve efficiency and productivity by identifying inefficiencies in workflows and adjusting them in real time.

For businesses facing rising labor costs and increasing demand for speed and personalization, this evolution is more than a technological advancement. It offers a strategic advantage.

Why ROI Measurement Is Different for Agentic AI

Traditional ROI models are rather straightforward as they compare the cost of a system to the output generated. ROI on projects using traditional models is measured based on cost savings, headcount reduction and cycle-time compression. However, agentic AI is more dynamic because the systems improve over time. This means the output isn’t static – rather, it compounds. These systems also reduce the need for ongoing supervision, operate continuously, and often uncover efficiencies that were not initially anticipated.

As a result, the ROI of agentic AI is not just immediate cost savings but also includes long-term gains. These gains include improved decision-making, faster execution, higher productivity, strategic agility and the ability to scale operations without a proportional increase in cost. Measuring this kind of value requires a broader, more forward-looking approach.

Key ROI Drivers of Agentic AI workflows

  1. Operational efficiency – unlike conventional automation that is vulnerable to dynamic environments due to fixed rules, agentic AI responds to changes automatically. These systems continuously learn and optimize, delivering ongoing improvements without additional manual effort.
  2. Real-time responsiveness – customers expect real-time interaction. Agentic workflows enable this through systems that are always on and context-aware.
  3. Scalability – businesses can handle increased demand without a corresponding increase in operational costs or headcount, allowing more efficient growth.
  4. Cross-departmental reach – Agentic AI agents can seamlessly connect workflows across different departments like HR, IT, and finance. This reduces operational friction between teams and enhances overall efficiency.
  5. Productivity gains – Agentic AI can operate 24/7, completing tasks faster and with greater consistency than human teams. This allows employees to focus on higher-value work, increasing overall organizational productivity.
  6. Cost reduction – by automating complex workflows, businesses can reduce reliance on manual labor, minimize errors, and eliminate inefficiencies. This can translate into significant savings.
  7. Revenue growth – Agentic AI enables faster go-to-market strategies and more personalized customer experiences. This can directly impact conversion rates and revenue.
  8. Improved decision quality – With access to real-time data and advanced analytics, agentic AI systems can make quick, informed decisions. This reduces human bias and enhances accuracy in areas like forecasting, inventory management, and customer engagement.

Strategies for Evaluating Agentic AI ROI

To measure agentic AI ROI, businesses need a structured approach that connects AI deployment to business outcomes.

  1. Identify high-impact workflows – repetitive, resource-heavy processes like IT support, sales operations, or compliance.
  2. Establish baseline measurements by documenting current costs, completion times, error rates, and headcount before deployment.
  3. Compare pre- and post-implementation performance by checking utilization rates, tasks completed, and infrastructure costs to confirm operational sustainability.
  4. Estimate agentic impact by projecting improvements in speed, cost, throughput, and quality.
  5. If implementing agentic AI in phases, use control groups to isolate its impact from other organizational changes.
  6. Measure real business outcomes, including cost reductions, revenue growth, and productivity gains.

Conclusion

Traditional automation delivered value by reducing manual effort. Agentic AI, on the other hand, reduces decision latency, operational friction, and coordination costs. Therefore, AI agents’ ROI is not defined by savings alone. Its real value lies in the ability to generate compounding returns across multiple dimensions of a business. By adopting a broader view of ROI, organizations can better assess impact, build stronger adoption cases, and identify new opportunities for optimization.