The Hyderabad edition of the SpeakIn Asia Dialogues Forum 2026, held on 19 and 20 January, opened with a paradox that shaped every conversation that followed. India has climbed to third place on Stanford's 2025 Global AI Vibrancy Index, ranks second globally for AI talent and has seen its AI workforce grow by 252% since 2016. According to SAP research, 93% of Indian businesses expect a positive return on AI within three years. Yet participants argued that the large majority of AI investments worldwide never deliver their promised value, and the data closer to home points to the same strain: 64% of organisations cite incomplete AI training as a barrier, and 48% describe their AI investments as fragmented.

The forum's diagnosis was that failure rarely comes from a lack of capability. It comes from misplaced intent. Boards demand AI because competitors have announced it, investors expect it, and customers ask for it, often when a simpler solution would serve better. Organisations respond by asking which technology to use rather than which problem to solve.

"Every model we were building had one significant gap: context. Technology without context creates no value. Nearly 90% of global AI investments are write-offs because of FOMO." Ritesh Dogra, Gen AI Innovation Head

Rupessh Goel, Managing Director and Head of Credit, Data and Analytics at First Citizens India, challenged the most common justification for AI spending. Too many AI conversations, he argued, focus on reducing cost. The better measure is decision velocity: how many decision points an organisation has been able to remove. India's GCCs in particular, he observed, have been chasing cost rather than reducing decision nodes.

A second thread reframed one of India's most criticised institutions. Many global CEOs came through India's imperfect higher education system. Participants argued that they thrived not because the system was flawless, but because they learned to think independently, adapt quickly and challenge assumptions. In an era when AI can process information faster and at greater scale than any person, critical thinking becomes the differentiating human capability. The forum described it as something built through three experiences: exposure to diverse perspectives, the discipline of questioning, and practical experience that develops judgement over time.

Dr. Padmakumar Nair, Vice Chancellor of Thapar Institute of Engineering and Technology, placed intellectual humility at the start of that process, noting that the human brain evolved to solve simple, dangerous problems quickly, while today's problems are complex and reward patience. Hari Chandana Dasari of the Indian Administrative Service called for education that teaches adaptability rather than training students toward a single predefined job. Ashish Babbar of Microsoft recalled a company blog post from his early years that ended with one line: we need people to think.

Sudipto Mandal, CHRO at L&T Metro Rail Hyderabad, warned of the opposite outcome.

"If human beings become just users rather than thinkers, we've failed. AI should enable us to think better, not replace our thinking." Sudipto Mandal, CHRO, L&T Metro Rail Hyderabad, Larsen & Toubro

What distinguished successful organisations, participants agreed, was not the sophistication of their technology but the clarity of their purpose. AV Ranganath, Commissioner of HYDRAA in the Telangana Government, drew a lesson from urban governance: certainty of action and speed of execution are what create impact, and organisations must act with the same clarity. The forum reduced this discipline to four questions every AI initiative should be able to answer. What problem are we solving? Are we driving revenue? Are we improving outcomes? Are we creating real, measurable impact?

Sector leaders brought the theme to their own industries. Technology companies, despite leading on readiness, still struggle to move from showcasing capability to demonstrating value. Financial institutions must balance automation with oversight and customer trust. Manufacturing shows the widest gap between potential and execution, while healthcare leaders stressed that AI should support clinical judgement rather than replace it.

The path forward, as the forum framed it, is to start with problems rather than technology, challenging every AI proposal with a simple test of whether a simpler solution would work better. It means building critical thinking deliberately alongside technical training, consolidating fragmented pilots into a focused enterprise strategy, and rewarding business impact rather than technology adoption. As Swetha Vissapragada of Electronic Arts observed, the biggest gains are not coming from automation itself, but from people finally having time to do real advisory work.

Technology enables scale. People create value. The organisations that join the successful minority will be those that resist FOMO and integrate technology with purpose.