In the near term, your existing enterprise platforms are not going away. Analysts consistently point out that core applications will remain in place for years, even decades, especially where they run mission-critical processes.

What will change is how you get value from those systems:

• Incumbent vendors are embedding agents directly into their platforms rather than replacing them. Microsoft, Oracle, SAP, Salesforce, and ServiceNow are all moving in this direction.
• Agentic AI will be rolled out gradually, with humans staying in the loop for a long time. There is still limited appetite in large enterprises to hand full autonomy to agents.
• You’ll see erosion around the edges of traditional apps (simple point tools, lightweight workflows), not wholesale rip-and-replace of your core CRM or ERP.

For CIOs, the practical takeaway is to optimize, not abandon, your current portfolio. Focus on:

• Adding agentic capabilities to existing platforms to increase ROI on multimillion-dollar investments.
• Identifying simple, generic point products that may be replaced by agentic workflows.
• Planning for an agentic orchestration layer that will sit on top of your current systems, rather than assuming those systems will disappear.

Agentic AI is expected to rethink how software is priced, shifting away from traditional seat-based subscriptions toward models tied to usage and outcomes.

Key trends to watch:

• From seats to outcomes: If an agent replaces a human task, customers will increasingly expect to pay for work done, not for log-ons. Vendors like Intercom and Salesforce are already moving in this direction.
• Decline of pure seat-based pricing: IDC predicts that by 2028, pure seat-based pricing will be obsolete, and 70% of software vendors will have refactored pricing around new value metrics such as consumption, outcomes, or organizational capability.
• Hybrid licensing models: Forrester’s view is that you may, for example, hold a subscription for 100 seats but swap 10–20 of those seats for consumption- or outcome-based pricing.

What this means for CIOs and IT finance:

• Use AI to lower total IT cost: AI agents can analyze consumption and usage patterns, giving you data to negotiate more favorable contracts.
• Expect new tiers and flex options: Vendors will introduce agentic tiers, flex bundles, and outcome-based offers. Consultants may even offer to manage agentic AI and charge a percentage of outcomes.
• Revisit your sourcing strategy: Build scenarios for how your spend profile changes as more work is done by agents rather than humans, and align contracts accordingly.

Agentic AI is set to reshape both your application architecture and the user experience, without necessarily eliminating your core systems.

On the application side:

• Blurring of CRM, ERP, and ITSM boundaries: Because agents don’t care which system data comes from, vendors are building unified data and workflow platforms that cut across categories.
• Examples include:
  • Oracle: integrated cloud ERP and CRM with a managed agentic platform.
  • Microsoft: ERP and CRM under Dynamics 365, plus industry-specific agentic offerings using smaller, cost-effective language models.
  • SAP: combining Signavio (process management), LeanIX (enterprise architecture), and its Joule AI agent into a cohesive system.
  • Salesforce: merging Mulesoft, Data360, and Agentforce into a more unified platform.
  • ServiceNow: using its acquisition of Moveworks to expand from ITSM into areas like CRM.
• Pressure on point products:
  • Simple, generic tools (basic workflow, spreadsheets, lightweight project apps) are at risk because they’re easy for agents to replicate.
  • Highly verticalized apps (e.g., Epic, Cerner, IQVIA, Procore) are better insulated due to deep domain expertise and integrations.
• Rise of vibe coding: End users can use natural language with tools like ChatGPT, Gemini, Claude, or GitHub Copilot to build their own apps and agents, especially for point solutions outside traditional platforms.

On the user experience side:

• Agentic layer as the new interface: Analysts expect the primary UI to become agent-driven and conversational, with traditional SaaS UIs increasingly hidden behind an orchestration layer.
• Reduced complexity: Instead of navigating multiple dashboards, users will ask an agent to perform tasks across systems, which can abstract away complexity and automate repetitive work.
• Orchestration across tools: An orchestration agent may route work to an LLM, an SLM, or RPA based on efficiency, cost, and even energy usage.

For CIOs, this points to a roadmap where you:

• Plan for a cross-platform agentic layer that can sit above CRM, ERP, ITSM, and other systems.
• Rationalize generic point tools that agents can replace, while protecting and enhancing high-value vertical applications.
• Define guardrails for vibe coding so business users can innovate safely without compromising mission-critical workflows.