Qlik recently issued a call to action from its AI Council on the shifts companies should be preparing for as AI moves deeper into decision support, workflow execution, and day-to-day operations.
The Council’s message is clear: the next phase of AI will be shaped by forces many organizations are still underestimating. Evaluation and accountability will carry more weight. Policy environments will keep fragmenting. Reasoning quality will face greater scrutiny. Model and interface churn will continue. Architectural choices will determine how quickly companies can adapt without repeated reinvention.
“AI is entering a harder, more consequential phase,” said Mike Capone, CEO, Qlik. “The easy conversations are behind us. Access to powerful models is widespread. The harder question is whether AI can operate inside the actual conditions of a business, with trusted data, accountable reasoning, evolving policy demands, and the flexibility to keep adapting as the market changes.”
Five voices on what companies should prepare for
“Many organizations still treat governance as a document set,” said Dr. Rumman Chowdhury, responsible AI leader, engineer, auditor, and investor. “That approach will fail under real pressure. As AI moves closer to decisions and actions, trust will depend on evidence. Evaluation needs to run continuously, under real conditions, with clear signals for when systems are reliable and when they are not.”
“The next AI divide will be shaped by power, access, and dependence,” said Nina Schick, author, advisor, and founder of an AI advisory firm. “Intelligence is being industrialized, concentrated, and contested at the same time. Leaders need to think beyond tooling decisions and focus on whether their organizations are built to adapt as the structure of the AI economy shifts.”
“Policy fragmentation is becoming an operating reality for global companies,” said Kelly Forbes, Co-Founder and Executive Director, AI Asia Pacific Institute. “Different markets are moving at different speeds, with different expectations around transparency, labor impact, oversight, and acceptable use. Companies that scale effectively will treat coordination and adaptability as core capabilities from the start.”
“A fluent output can still reflect shallow reasoning,” said Michael Bronstein, DeepMind Professor of Artificial Intelligence, University of Oxford. “The systems that matter in business will be the ones that can work with structure, relationships, and constraints. Context is what makes intelligence useful inside a real organization.”
“The model layer is going to keep changing faster than most enterprise planning cycles,” said Mark Relph, Director – Data and AI Go-To-Market (GTM) – AWS. “Companies should assume new models, new assistants, and new orchestration patterns will keep arriving. The durable choice is to stay open, governed, and ready to adopt what works without reworking the whole system each time.”
Taken together, the council’s perspective points to a more demanding standard for AI readiness. Companies will need systems that stand up to scrutiny, operate with trusted context, absorb better models as they emerge, and stay useful as business, regulatory, and technical conditions keep shifting.
Leave a comment