The Agentic AI Paradox: Why 75% Use AI, But 95% of T&E Reporting Projects Still Fail

The T&E AI Commitment Gap: Why 75% Use AI, But Most Projects Fail

Last week at Business Travel Show America (BTSA) 2025, the energy around Artificial Intelligence was palpable. During the "AI in Action: How Travel Managers Are Engineering Smarter Programs" panel, moderator Chris Davis (Managing Editor, The BTN Group, surveyed the audience, finding that 75% of attendees reported using AI in their operations.

This high adoption rate is a clear sign of ambition within the corporate travel management sector. However, as Keesup Choe (CEO, PredictX ), immediately countered, that figure represents only part of the story. Citing a new study by MIT on corporate initiatives, Choe warned the industry that "95% of corporate AI initiatives fail".

This staggering statistic, as reported by PhocusWire, perfectly defines the Enterprise AI Paradox that we explored in our recent series: widespread investment and adoption drastically outweigh measurable success. The core lesson from the panel, which included experts like Andres Fabris (Founder & CEO, Traxo) and Jennifer Steinke (Director Travel, Meetings & Fleet, Moderna), is simple: Use doesn't equal success; commitment and strategy matter.

Strategic Flaw: Moving Beyond Chatbots to True T&E Predictive Analytics

The panel discussion quickly pivoted from general adoption to the strategic nuance required for success — the very nuance missing from the 95% of failed projects.

Andres Fabris emphasized that we miss crucial insights if we view AI's impact monolithically. He warned that most adoption today is occurring at the entry-level "chatbot level". While this is a "nice point of entry, you can't stop there," he stressed.

This strategic oversight aligns perfectly with our previous analysis that AI projects fail from a lack of strategic alignment (Part 2 of our series). Keesup Choe reinforced this idea, stating that the process must "start with humans." Leaders must have a clear vision of where they want to be, "and then ask how AI can help you" achieve that strategy.

The PredictX Platform: Solving the T&E Data Fragmentation Crisis

The strategic vision required for the 5% of successful programs fundamentally depends on solving the data fragmentation that plagues corporate travel. The high-level role of the PredictX platform is to eliminate this foundational failure, turning fragmented sources into reliable, actionable intelligence.

1. Achieving Data Consolidation & Insights: The PredictX platform is purpose-built to ingest data from over 200 connectors (TMCs, expense systems, corporate cards) to create a single consolidated T&E data source. This foundational step ensures that when a travel manager asks a strategic question, the Agentic AI is trained on 99.9% accurate, contextualized data, avoiding the "garbage in, garbage out" trap that derails 70% of AI initiatives.
2. Enforcing Governance and Strategy: By unifying data, the platform provides the high-level governance necessary to mitigate legal and financial risk. It acts as the backbone for compliance, guaranteeing that AI decisions and T&E Reporting are based on governed data, moving beyond the reactive, post-audit world that exposed Air Canada to liability.

Cogent Agentic AI: Engineering Smarter T&E Programs with Actionable Insights

Cogent, our award-winning Agentic AI workspace, is the powerful, actionable solution that embodies the strategic commitment discussed by the BTSA panelists. It moves beyond simple chatbots to function as a goal-oriented virtual workforce.

1. Driving Proactive Predictive Analytics: Cogent is the tool that transforms the concept of a "learned skillset" into reality. It enables T&E professionals with conversational intelligence, allowing users to ask complex questions in natural language and instantly generate dynamic, customized reports and Travel Data & Predictive Analytics. This capability allows buyers to monitor KPIs autonomously, surface savings opportunities invisible to the naked eye, and shift the T&E role from data reporter to strategic leader.
2. Operationalizing Strategic Goals: Cogent's Agentic AI provides the necessary commitment for success by managing the full lifecycle of complex processes. It acts as an intelligent workforce of AI agents that can analyze vendor performance for strong negotiations, manage the full air contract lifecycle, and accelerate RFP analysis with unmatched speed. This level of focused, multi-step automation is precisely the "action" required to achieve the strategic vision that Keesup Choe identified as the marker of the 5% who succeed.

Join the 5%: Your Blueprint for T&E Reporting Success

The message from Business Travel Show America is clear: The competitive edge in corporate travel now belongs to organizations that build their AI programs with a human-led vision and strategic commitment. Are you ready to move your Travel & Expense Reporting beyond the 95% failure rate?

Master Your Data: Read the Full T&E Success Blueprint Series

For a deeper dive into the AI Paradox and the steps required to join the 5% who succeed, read our three-part blog series:

• Part 1: The Foundation: The AI Failure Epidemic: Why McDonald's & Air Canada's Failures Reveal a T&E Reporting Crisis
• Part 2: The Analysis: Beyond the Code: The Strategic and Data Flaws Ensuring Your AI Project Fails
• Part 3: The Solution: The Agentic AI Revolution: How to Join the 5% Who Succeed by Investing in People

Transform your T&E management today. Request a trial of Cogent today.

Unlock Your T&E Advantage: Subscribe to PredictX in Focus Newsletter

🔔 Stay Ahead! Follow Us on LinkedIn for the Latest Innovations and Feature Updates to Enhance Your Travel Management Experience!

‍