The State of AI in 2025: What’s Really Happening Inside Organizations

AI isn’t quite where we thought it would be.

In this new blog series, we’ll explore the real-world state of AI; how organizations are planning, piloting, deploying, managing, securing, and adopting AI tools like Microsoft Copilot. We’ll break down what’s working, what’s not, and how to ensure your AI investment delivers measurable value.

For this first post, I want to set the stage. Over the past six months, customer priorities around AI have shifted dramatically. Let’s start with a simple definition of AI from my perspective:

AI is a tool that enables employees to focus on the skills they were hired for by automating and executing time-consuming or repetitive tasks.

It’s not magic. It’s not a silver bullet. It’s a productivity multiplier when implemented correctly.

Four Core AI Use Cases in 2025

Across industries, organizations are focusing AI investments in four main areas:

• AI Content Creation – Generating emails, reports, research summaries, analytics, and contracts.
• AI Content Discovery – Advanced search and data analytics.
• AI Simulations – Modeling in medical, financial, and manufacturing sectors.
• Conversational AI – Assistants, chatbots, and automated help tools.

Now, here’s the reality: the average Fortune 500 company with 10,000 employees spends roughly $3.6 million per year on AI tools like Microsoft Copilot. That’s a major investment. To justify it, most executives expect $5–9 million in return, and if they don’t see it, AI budgets get scrutinized fast.

Shifting AI Priorities in 2025 

What do I mean? Let's look at the changes in priorities around AI this year alone. Between January and July 2025, global policies and economic uncertainty seemed to have triggered a major pivot in AI strategy. 

January 2025 – Growth & Expansion 

• 90% of companies planned to invest $1M–$5M in AI (Forbes) 

• 66% hired staff specifically for AI implementation (WSJ) 

• 28% reported AI had already helped cut costs (Gartner) 

July 2025 – ROI & Cost Control 

• Only 25% of AI projects met ROI expectations (IBM CEO Study) 

• 41% planned workforce reductions where AI could replace tasks (World Economic Forum) 

• 30% were reallocating budgets to AI by cutting salaries, benefits, and office space (Gartner) 

The focus has shifted from “grow at all costs” to“make AI pay for itself.” 

Four Areas Where Microsoft Copilot Rollouts
Deliver Value 

As a result of these shifts, we’re seeing CEOs concentrating on four high-impact areas to maximize AI ROI: 

1. Strategic integration of Generative AI. 

• Where does AI truly add value? One example is using AI to build a custom HR chatbot. Employees can easily find answers about parental leave policies, remaining vacation days, or sick time, without needing direct access to secure HR systems. By pairing AI with targeted, approved data, companies can improve information flow while protecting sensitive information. 

2. Enhanced utilization and optimization of existing technologies. 

• Leaders are investing in training employees to fully adopt tools like Copilot, moving beyond basic use to creating custom prompts, agents, and integrated workflows. By connecting Copilot to third-party apps such as Jira, Asana, Trello, and Slack through Copilot Connectors, organizations can stack technologies and AI capabilities to automate more complex productivity processes. 

3. Deeper focus on practical value. 

• The conversation around AI is shifting from hype to tangible outcomes and AI impact. Leaders are encouraging teams to identify time-consuming, repetitive tasks that AI can handle, freeing employees to focus on strategic, creative, and high-value work that aligns with their core strengths and job purpose. 

4. Increase use of customer-facing AI. 

• Custom agents/chatbots can be quickly built by connecting AI to a clean, curated data source containing the most common customer questions. This approach can significantly reduce Tier 1 and Tier 2 support calls, improving response times and allowing support teams to focus on more complex issues. (I’ll explore how to set this up in a future post.) 

Why 1 in 3 AI Rollouts Stall and How to Avoid It 

From January 2024 to March 2025, one in three organizations paused or abandoned their AI rollouts. The primary reason? They skipped a critical step in AI readiness: data cleanup. 

Rolling out an AI tool like Microsoft Copilot without first cleaning, organizing, and archiving outdated data almost guarantees poor results. When your AI relies on outdated, inaccurate, or irrelevant information, users will quickly lose trust in its responses. Bad or outdated answers lead to low adoption, and in many cases, complete abandonment of the tool. 

In future articles, we’ll dive deeper into how to prepare your data for AI success. For now, remember: every organization should be actively maintaining a data housekeeping cycle to keep AI outputs accurate, relevant, and reliable. 

What’s Next in This Series 

Over the next few blogs, we’ll take a closer look at each stage of a successful Copilot rollout: plan, pilot, deploy, manage, secure, and adopt. Along the way, I’ll share:  

• Why some customers aren’t seeing value from Microsoft Copilot 

• Practical ways to avoid common mistakes in your Copilot journey 

• Real-world tools, tips, and success stories from AI adopters 

I also want to hear from you! What’s worked in your AI journey? What hasn’t? Share your lessons learned, and I may feature them in a future post.  

Want to learn more? You can attend my sessions and meet with me at the following conferences: 

• TechCon365-  Atlanta, GA  – August 11-13 
• TechMentor-  Redmond, WA – August 14-15 
• TechCon365- Irvine, TX -  Nov 3-7 

-Stephen 

Transform the way your organization adopts Microsoft 365 and Copilot with ENow True Adoption Center

Instead of leaving employees to "figure it out" or relying on outdated, generic training, ENow gives you actionable insights into real adoption and usage trends across your tenant. Identify gaps in digital literacy, understand how teams are truly working, and design evidence-based adoption programs that drive measurable behavior change. With True Adoption Center, you’ll gain the clarity and data you need to maximize Microsoft 365 and Copilot ROI, turning technology investments into real productivity gains. Discover how ENow can help you take control of adoption today.