A Story of Two Futures

A BCG's 2025 "Build for the Future" study, supported by McKinsey and IBM research

Artificial Intelligence is already a ‘daily life’ in today’s era. It has become a boardroom priority, a competitive pressure, and a major investment agenda across industries.

In 2024 alone, corporate AI investment reached US$252.3 billion, according to Stanford HAI’s 2025 AI Index Report. Yet the real question is: are they actually generating value from it?

The answer, for many organizations, is still disappointing.

A 2025 BCG Build for the Future study of more than 1,250 companies found that only 5% of companies are achieving AI value at scale. Meanwhile, 60% are not achieving material value at all, despite substantial investment, and another 35% are scaling AI but still not moving far enough or fast enough.

The Problem: Many Companies Are Stuck in “Pilot Mode”

Across industries, many companies have launched AI initiatives, formed AI task forces, tested generative AI tools, and experimented with automation use cases. However, experimentation does not always translate into enterprise-wide impact.

McKinsey’s 2025 global AI survey found that nearly nine out of ten organizations are regularly using AI, but most have not embedded AI deeply enough into their workflows and processes to capture material enterprise-level value. McKinsey also found that AI high performers represent only around 6% of respondents.

This means that AI adoption is becoming common, but AI value creation is still rare.

The 5% Are Pulling Further Ahead

The companies that are winning with AI are accelerating. BCG calls these companies “future-built.” They have developed the capabilities needed to make AI work not only for productivity, but also for innovation, reinvention, and business growth. According to BCG, these future-built companies achieve five times the revenue increases and three times the cost reductions from AI compared with other companies. They also demonstrate stronger shareholder return performance.

The reason is simple: early AI wins create a compounding effect.

BCG’s research shows that future-built companies plan to spend 26% more on IT, dedicate up to 64% more of their IT budget to AI, and invest 120% more overall in AI than laggards. This is how the AI gap becomes harder to close. They are building a cycle of value creation that allows them to move faster every year.

The 10-20-70 Rule

BCG’s 10-20-70 rule suggests only 10% of AI value comes from algorithms, 20% from technology and data, and the remaining 70% from people, processes, and change management.

Companies often focus heavily on the tool, but underestimate the organizational change required to make the tool useful. Which means, they deploy AI on top of old processes.

Future-built companies do something different. They ask how would the process looks like if they build it from scratch with AI available from the beginning. Hence why, the value of AI comes from reshaping workflows end-to-end.

The Next Wave: Agentic AI

Just as many companies are still trying to scale generative AI, the next wave is already emerging: agentic AI. Agentic AI can reason, plan, act, coordinate, and adapt toward a goal with less human initiation.

BCG reports that AI agents already account for around 17% of total AI value in 2025 and are expected to reach 29% by 2028.

McKinsey also found strong interest in AI agents, with 62% of survey respondents saying their organizations are at least experimenting with them. However, many organizations remain in the early stages of scaling AI across the enterprise.

This means the competitive window is narrowing.

A Practical Roadmap for the 95%

For companies that are not yet generating value from AI at scale, the path forward should not begin with more random pilots. Here are six stages of practical roadmap of AI transformation:

1. Face Reality

Map all existing AI initiatives. Identify which ones are creating measurable business value, which ones are still experimental, and which ones should be stopped.

2. Choose the Right Battleground

Focus on two or three core business functions where AI can create significant impact. Avoid spreading resources too thin across too many disconnected use cases.

3. Build the Foundation

Strengthen data quality, governance, system integration, and platform architecture. AI cannot scale effectively when data and technology remain fragmented.

4. Redesign the Workflow

Do not simply deploy AI into existing processes. Redesign how work gets done, how decisions are made, and how humans and AI collaborate.

5. Develop People and Capabilities

Train employees, managers, and leaders to use AI effectively. The AI skills gap is not only a workforce issue; it is also a leadership one. BCG notes that companies often focus too much on launching AI solutions and not enough on ensuring that people can use them meaningfully in daily work.

6. Scale and Reinforce

Document successful use cases, build reusable AI components, track measurable outcomes, and reinvest the gains into stronger capabilities.

From AI Awareness to AI Execution: How Multimatics Helps Organizations Move Forward

The AI value gap shows that organizations need the right capability, workflow, governance, and implementation roadmap. Hence, this is where structured AI capability development becomes critical.

To help organizations move beyond experimentation, Multimatics designed these programs to empower teams to work smarter, scale faster, and apply AI in real business environments.

1. AXIS: Agentic AI for Enterprise Systems

AXIS helps organizations understand and build autonomous AI agents that can support decision-making, execute tasks, and coordinate actions across enterprise systems.

This program is suitable for organizations that want to explore how agentic AI can be applied beyond simple prompting or chatbot usage. Participants will learn how to design AI agents from the foundation stage, build and orchestrate them, and connect them into enterprise-level use cases.

Program phases:

• Phase 1: Foundation
• Phase 2: Building & Orchestrating
• Phase 3: Enterprise

2. FlowMind: AI Automation & Intelligent Workflow

FlowMind focuses on helping organizations automate routine and repetitive tasks through AI-powered workflows.

This program is relevant for teams that handle operational processes such as data processing, scheduling, reporting, ticket management, approvals, and other workflow-heavy activities. Instead of only using AI as a productivity tool, FlowMind helps teams redesign work processes so AI can support efficiency at scale.

Program phases:

• Phase 1: Process Analysis & RPA
• Phase 2: AI-Enhanced Automation
• Phase 3: Orchestration & Governance

3. MARS: Multimatics Advanced RAG System

MARS is designed for organizations that want to build Retrieval-Augmented Generation systems for business and enterprise environments.

Many organizations already have valuable knowledge stored in documents, policies, reports, manuals, training materials, and internal databases. MARS helps teams turn that knowledge into AI-powered systems that can retrieve, contextualize, and generate more relevant responses based on enterprise information.

Program phases include:

• Foundation & Knowledge Base
• Pipeline & Interface
• MVP + Demo Day
• Post-Training Coaching

Through these programs, Multimatics helps organizations build the practical AI capabilities needed to move from isolated pilots to scalable transformation.

Ready to move from AI experimentation to real AI transformation?

Explore Multimatics Artificial Intelligence Program and discover how AXIS, FlowMind, and MARS can help your organization work smarter, scale faster, and build future-ready AI capabilities.