This Month in AI - December 2025

​Welcome to the December 2025 edition of "This Month in AI." Here you'll find an overview of the latest articles and reports on advancements in artificial intelligence, trends in enterprise adoption, and innovative strategies business leaders like you are using to harness AI's transformative power.

​Here’s what’s making news this month:

‍

​The New AI Frontier: Frontline Work as the Next Competitive Divide

Artificial intelligence is no longer just a back‑office or knowledge-worker phenomenon, it is poised to transform the backbone of the global economy: frontline workers who directly serve the customer.

According to Why frontline work is AI’s biggest opportunity by The People Space, frontline employees comprise roughly 80 percent of the workforce, yet they have received a fraction of HR tech investment to date. That imbalance is both the cause of persistent operational instability and the source of AI’s most promising opportunity.

Frontline industries from retail and healthcare to logistics and hospitality are struggling with foundational barriers that degrade worker experience and operational outcomes. Issues like volatile scheduling, fragmented systems, under-resourced supervisors, and tools designed for office workers, not mobile shift workers, have eroded productivity and morale. Turnover often gets labeled the “frontline crisis,” but deeper structural friction, not pay alone, is the true culprit.

Why AI Matters Most at the Frontline

Contrary to the office-centric AI narrative dominating much of enterprise discourse, data suggests the highest ROI for AI is on the frontline. Intelligent systems can:

• Automate and optimize real-time scheduling and overtime allocation.
• Provide mobile-based HR support for everything from policy questions to time-off requests.
• Unite disparate systems as scheduling, timekeeping, pay into a coherent operational backbone.
• Support faster onboarding, contextual learning and development nudges, and continuous coaching.

This shift reframes AI from a standalone technology project to an operational imperative co-owned by HR and operations, designed around unit-level outcomes rather than desktop-based workflows.

Closing the Readiness Gap

Despite this potential, frontline industries currently exhibit some of the lowest AI readiness of any sector. Legacy HR systems were built for knowledge workers; frontline employees often still rely on shared terminals, static noticeboards, or siloed apps with inconsistent data formats. Without unifying these foundational systems, high-value AI use cases simply cannot scale. Furthermore, supervisors who are central to frontline performance remain under-equipped with digital tools and training.

What Future-Ready Frontline AI Looks Like

Imagine every shift supervisor armed with an AI copilot on their mobile device. This assistant would:

• Provide instant policy guidance and staffing insights.
• Auto-generate recognition messages and coaching prompts.
• Help balance labor demand with employee preferences in real time.

Workers, meanwhile, get fast, conversational access to answers about pay, leave, and scheduling removing delays and reducing churn. These capabilities together can elevate operational reliability: fewer call-outs, better shift coverage, and more predictable labor costs.

The Cost of Office-First AI

Failing to prioritize a frontline-first strategy risks reinforcing a two-tier workforce. Office employees will benefit from productivity tools while frontline teams remain stuck with retrofitted, ineffective systems. The cultural divide deepens, supervisors fall further behind, and operational instability grows - ultimately degrading service quality and customer experience.

Blueprint for Action

HR and organizational leaders must break from office-centric thinking and:

• Build mobile and voice-first tools that meet workers where they are.
• Fully integrate scheduling, time, attendance, and pay data.
• Start AI enablement with supervisors as the linchpin for consistent decision-making.
• Prioritize AI touchpoints that deliver measurable value in minutes.
• Co-own frontline AI use cases with operations rather than treat them as HR side projects.

A New Competitive Divide

Organizations willing to redesign systems around the needs of frontline work, rather than layer new tools onto outdated processes, will stabilize teams, strengthen customer experiences, and build resilience into the most critical part of their workforce.

How Mentor Mindset’s AI app for QSR managers cut employee attrition dramatically → Read the case study

‍

What Agentic AI Means for the Banking Frontline

The next wave of AI impact in banking will be won or lost at the point of customer engagement. According to McKinsey’s Agentic AI is here. Is your bank’s frontline team ready?, agentic AI, a system that can autonomously plan and execute multi-step tasks, is beginning to reshape how relationship managers sell, serve, and advise. Rather than simply assisting with individual tasks, these systems orchestrate end-to-end workflows, positioning the banking frontline as the next major lever for growth and productivity.

Why “AI is good for what ails you”

The banking frontline sits at the intersection of revenue, trust, and customer experience, yet it remains weighted down by fragmented systems and manual coordination. Relationship managers often spend more time navigating internal processes than engaging with clients.

Agentic AI directly targets this friction. By automating orchestration across systems, it frees frontline teams to focus on judgment, relationships, and complex decision-making. McKinsey highlights this as one of the highest-impact AI opportunities in banking, particularly in commercial and corporate segments where complexity and deal velocity are high. More than three quarters of the leaders surveyed have great expectations for the technology

Early Movers vs. the Rest

As with broader enterprise AI adoption, a performance gap is already emerging. The most forward thinking banks  are moving aggressively - piloting agentic workflows, redesigning frontline roles, and aligning technology with business outcomes rather than experimentation.

These leaders share several characteristics:

• They redesign end-to-end client journeys instead of layering AI onto existing tasks.  
• They embed agentic AI directly into frontline workflows, not separate tools.  
• They invest in strong governance to manage risk, accountability, and escalation.  
• They treat AI as a growth lever, not just an efficiency play.  

By contrast, most institutions remain stuck in pilot mode, testing isolated use cases without rethinking operating models. The result is incremental efficiency gains, but limited strategic impact.

The Readiness Gap

Despite the promise, McKinsey highlights a critical constraint: most frontline teams are not yet ready for agentic AI at scale. Key blockers include fragmented data architectures, inconsistent CRM adoption, unclear decision rights between humans and machines, and unresolved questions around accountability and regulatory compliance. In highly regulated environments like banking, autonomy without guardrails is not an option.

Equally important is change management. Frontline adoption hinges on trust that AI recommendations are accurate, explainable, and aligned with client interests. Without this, agentic systems risk being ignored or actively resisted.

What a Future-Ready Frontline Looks Like

In a mature agentic AI model, frontline bankers operate with an AI copilot that works ahead of them, not just alongside them.

Before a client meeting, the agent assembles relevant financials, flags risks and opportunities, and proposes tailored talking points. After the meeting, it drafts follow-ups, updates systems, and coordinates next steps across internal teams. Managers gain real-time visibility into pipeline health and coaching opportunities, rather than retrospective reports.

Importantly, humans remain in control. Agentic AI handles orchestration and execution; bankers provide judgment, relationship insight, and accountability.

The Cost of Inaction

Banks that delay frontline transformation risk falling into a familiar pattern: investing heavily in AI infrastructure while failing to translate it into frontline impact. This creates a widening gap between AI-enabled institutions and those weighed down by operational drag.

Over time, this gap shows up in slower growth, inconsistent client experiences, and higher frontline burnout that no amount of back-office optimization can offset.

A Strategic Imperative for Banking Leaders

McKinsey’s conclusion is clear: agentic AI is no longer a future concept. It is already reshaping how leading banks operate at the frontline. The winners will be institutions that move beyond experimentation and treat agentic AI as an operating-model transformation that aligns technology, governance, and talent around frontline value creation.

In 2026, AI may be table stakes in banking. But competitive advantage will belong to the banks that make their frontline teams truly agent-ready.

‍

​AI as an Economic Actor, Not Just a Tool

Artificial intelligence is crossing a threshold most leaders haven’t fully grasped. Aithority’s article AI is becoming an economic actor, not just a tool highlights how AI systems have shifted from being passive helpers to entities that routinely make and execute economic decisions such as changing pricing, allocating resources, and shaping outcomes with or without human sign-off.

Beyond Optimization: AI With Decision Authority

For years, software enhanced human decision-making: dashboards summarized data, models suggested options, and analytics informed strategy. But as models improve and automation spreads, AI now routinely performs actions that affect value creation and distribution. Algorithms set credit limits, adjust real-time prices, divert supply-chain orders, and reallocate budgets based on performance signals, often without explicit human approval.

This isn’t incremental automation; it’s a structural shift in economic agency. Systems that once needed human commands now operate autonomously within set goals, subtly rewiring how decisions are made with humans and machines executing them.

The Economy is being Rewritten at Machine Speed

The transition from “decision support” to decision authority didn’t happen overnight. It accelerated as feedback loops, bigger data pipelines, and optimization logic steadily expanded AI’s scope. What was once labeled “efficiency improvement” now looks like non-human actors influencing market dynamics  like shaping incentives, reallocating resources, and even affecting employment and capital flows across industries.

This evolution reframes AI from infrastructure to a participant in economic activity. Unlike traditional tools, modern AI operates continuously, interpreting real-time signals and acting on them often before humans intervene.

Leadership and Governance in the Age of AI Actors

If AI systems now influence pricing, labor allocation, procurement, and internal capital flows, then business leaders must rethink governance, responsibility, and accountability accordingly. Seeing AI as “just software” leads to an oversight gap, especially when AI decisions affect who gets funded, hired, or prioritized within the organization.

The most pressing question for executives is no longer whether AI can make better choices, but whether leadership understands how far AI has already integrated into economic decision loops and what that implies for strategy, control, and value distribution.

A New Economic Framework Emerges

AI’s growing agency calls for new frameworks for accountability, governance, and economic participation. An AI actor doesn’t think or intend in human terms, but its outputs still affect value flows, incentives, and societal outcomes. Leaders must ensure that ethical constraints, transparency, and oversight are built into systems that now play a non-trivial role in economic activity. Seeing AI as a participant and not a passive assistant is no longer theoretical. It is essential for understanding how decisions are now made, who holds economic authority, and how future markets will function.

‍

Automation Isn’t Enough for Customer Success​

​According to Bain’s Customer Success at a Crossroads: Evolve with AI or Fade Away, the customer success function long hailed as the backbone of retention and expansion is losing momentum even as investment grows. While most leaders recognize AI’s potential to make retention efforts more efficient and effective, the majority remain stuck in pilots or narrow, low-impact use cases, and net revenue retention continues to decline in many software organizations.

​Stuck in Pilots, Not Impact

Bain’s research finds that while most customer success leaders grasp the promise of AI from automating routine outreach to preparing CSMs for calls and drafting success plans, about 70% have yet to scale meaningful AI use cases beyond experimentation.

This limited uptake leaves teams chasing micro-efficiencies rather than changing how work gets done. Root causes include treating AI like an IT project instead of a strategic priority and layering it onto legacy, human-centric workflows instead of redesigning them.

Where Value Really Lives

Customer success managers spend roughly two-thirds of their time on lower-value activities that could be automated, with the real opportunity lying in freeing them to build relationships, drive outcomes, and support expansion and renewal. AI can change this, but only if its deployment is guided by clarity and focus, not scattershot experimentation. Bain outlines four actions that separate organizations realizing real returns from those stagnating in pilots:

• Set bold ambition with clear targets and define what success means (e.g., higher net retention, more strategic time with customers).  
• Prioritize a few high-impact processes and avoid applying AI everywhere at once.  
• Redesign from first principles and rethink workflows with AI capabilities at the core, not as an add-on.
• Embed AI deeply into work and make new tools the default way of working to drive adoption.

The Strategic Imperative

Customer success still matters more than ever, but stagnant outcomes risk undermining the function’s credibility and impact. Teams that lean into AI as part of a fundamental reimagining of how they deliver value instead of just looking to automate tasks are the ones poised to reverse retention trends and redefine what customer success delivers at scale.

AI for customer success must move from pilot mode to purposeful, scaled impact, or the teams entrusted with customer growth risk fading from strategic relevance.

‍

A​gentic AI Is Transforming Supply Planning

​World Economic Forum’s article How agentic AI will change supply planning from firefighting to foresight, presents how supply planners today operate amid unprecedented volatility from trade disruptions and climate impact to regulatory shifts, making traditional planning methods increasingly inadequate. In this new reality, agentic AI is emerging not just as a tool, but as a strategic partner that helps planners navigate complexity with speed and precision.

From Manual Planning to Network-Level Insight

Supply planning has historically relied on individual site data and iterative manual processes. Experts now advocate a shift toward a holistic, network-centric planning perspective that balances coordinating plants, inventories, and capacities across a dynamic global footprint. This network logic, while more resilient, also creates data complexity that traditional spreadsheets and experience alone can’t manage.

Intelligence Where It Matters Most

AI-driven planning tools integrate diverse data from demand forecasts and inventory levels to labour availability and geopolitical risk, enabling scenario simulation and optimization across the entire network. These systems operate like real-time navigators, continuously scanning changing conditions and recommending optimized responses. Planners can model disruptions (e.g., port closures or sudden demand shifts) and immediately assess impacts — turning hours of manual analysis into minutes of insight.

What Agentic AI Brings to the Table

While today’s advanced planning systems are already a leap forward, the next frontier is agentic AI — autonomous agents that don’t wait for prompts but actively monitor real-time signals and propose adjustments before planners open their tools. These agents act as intelligent copilots, interpreting results, triggering alternative scenarios, and suggesting actionable responses based on evolving conditions.

Human Planners Remain Critical

Importantly, AI doesn’t replace humans but empowers them. Agentic systems automate routine reconciling and scenario generation, but planners still lead strategic orchestration, reconciling competing goals like cost, delivery performance, sustainability, and resilience. This partnership accelerates decision-making and frees planners to focus on stakeholder coordination and higher-order strategy.

Why It Matters Now

In the face of constant disruption, organizations that reimagine planning work with agentic AI stand to convert volatility into competitive advantage — shifting from reactive firefighting to anticipatory foresight that strengthens resilience, efficiency and collaboration across global supply networks.

‍

​That’s it for the December’s edition of This Month In AI. We hope you enjoyed the read.

​Sign up for our monthly newsletter to get this curated list of AI articles and more AI insights from RapidCanvas delivered straight to your inbox.