AI ROI Is Elusive Because You’re Building on Sand, Not First Principles

Posted on by

(SeaPRwire) –

By: Oliver Hawthorne

Most companies pour cash into AI and see zero meaningful returns. They’re obsessed with looking AI-savvy instead of solving real problems. This gap between hype and results has become the industry’s most pressing anxiety.

Last week at Brainstorm Tech in Aspen, AI dominated every conversation. At an Eye on AI breakfast, leaders laid bare root causes. State Street’s CTO Manoj Bohra said regulated firms need years of foundational work. This includes organizing data, mapping workflows, setting up governance. He compared it to building a bridge. No one judges its ROI in a single year. Deloitte’s Bill Briggs noted many firms drop AI into old processes. This weaponizes inefficiencies at scale. ReviveHealth’s Kathy Pham said companies often optimize for the wrong goals. For example, using AI for bedtime stories when the point is parent-child time. Lambda’s Stephen Balaban warned AI isn’t ready for most use cases outside software development. But firms should prepare for next year’s advances. Genspark’s Wen Sang advised chasing easy wins. Ad firms use AI for video prototypes to win more pitches at lower cost. Wells Fargo’s Faraz Shafiq shared they built cross-business AI blocks. These include a unified agent platform and governance infrastructure. They saw a 25% jump in new accounts. They also gained long-term customer goodwill from personalized service. Beyond the breakfast, Anthropic’s Boris Cherny emphasized in-person team time. His team supervises hundreds of AI agents in parallel. Hyatt and Snowflake’s CEOs slammed rivals restricting third-party AI agents. They called it weak toll-gating that customers won’t tolerate. Mistral’s Timothee Lacroix said AI sovereignty means controlling stack parts like data centers. The firm has no plans to build its own AI chips. Countries must protect indigenous data from exploitation, he and Defined.ai’s Daniela Braga agreed.

The commercial loop will sort winners from losers. Firms that skip foundational work will keep wasting money on vanity projects. Those that reengineer workflows and align AI with core purposes will capture sustainable ROI. Service firms will shift to outcome-based pricing as businesses demand results over headcount. Regional players will double down on controlled stack components to claim AI sovereignty. The era of AI hype is ending—only disciplined builders will survive.

Author bio: Oliver Hawthorne, Principal Correspondent at an international tech review, covers enterprise AI and global tech strategy from Silicon Valley.