Agentic AI and reflections from last week

Agentic AI and reflections from last week

Last week I managed to catch the Dreamforce 2024 event in San Francisco, immediately followed by a large enterprise CIO Summit in San Diego. It was a good way to reflect on the balance between the promise of tomorrow and the practical realities of today - and the critical need to get Data and People right. Here are my three takeaways from last week.

1. The Coming of Agentic AI

Marc Benioff introduced their agentic technology that can autonomously handle business-specific tasks like planning, decision-making, and reasoning. This moves us from an era of "AI copilots," into full-fledged "AI pilots" - AI that can take meaningful actions on behalf of users (in few immediate applications in customer-facing areas; broader use cases depending on the complexity of workflows and AI maturity to come). Satya Nadella echoed this transformation by stating that “every company will be an AI company,” emphasizing that businesses must adopt AI at their core—not just as a tool, but as a strategic enabler that drives organizational change. Jensen Huang added the next decade of AI innovation will be like watching "the movie of a lifetime" - a reminder that AI’s transformative power is only just beginning, and businesses that fail to embrace it risk being left behind.

2. The Critical Need for a Strong Data Foundation

But one of the most consistent themes over the week was the importance of data quality in driving AI success - in that AI is only as strong as your weakest data source. Generative AI, while powerful, is deeply reliant on the data it's trained on - underscoring the need for businesses to invest in robust data governance frameworks to deliver clean, well-governed data. Without strong data architectures, even the most sophisticated AI models will yield flawed or unreliable results, and AI’s real power comes from its ability to turn data into actionable outcomes across business functions. Its critical that businesses view their data not just as an asset, but as the backbone for future AI innovations.

3. Human and AI: A Symbiotic Future

AI isn't here to replace human workers but to enhance their capabilities, and yet, the journey needs to be navigated very carefully. Indeed, automation isn't about replacing jobs—it's about unlocking human potential - by automating mundane tasks, AI allows employees to focus on higher-level, creative problem-solving. This shift will enable businesses to unlock greater productivity and innovation, and it's getting clearer that AI will be as integral to our daily lives and work, as smartphones are today. Equally, AI's rapid adoption will lead to job transitions and workforce impact across sectors, and to address this companies must take an active role in reskilling and upskilling their workforce - indeed, organizations that proactively support employee transitions will position themselves as both competitive and socially responsible. Balancing technological advancement with human capital investment will be the key to navigating this new AI era responsibly.



Anupam Gupta

CPO @ Applied | Vertical SaaS Product Leader

3mo

Right on, Sanjay Srivastava ! Thanks for sharing.

Raghunath K.

Advisory to C-Suite | People Advisory Services | Human-centric Business Strategy & Transformation | Pre Sales | AI Practitioner | Data Analysis & Visualization | Strategic Planning & Execution

3mo

Thanks for sharing this Sanjay Insightful

Melville Carrie

Digital | Product | Data | Ai | Fellow | Views my own

3mo

Once again Sanjay Srivastava - as the saying goes... "you are on the money"! It makes total logical sense that if a series of interactions with a human are already codified - bound - and therefore limited, either by regulation, internal policy or indeed practice / procedure, e.g. within a banking environment - then it stands to reason that today's human-to-human interaction can likely be successfully replaced with human-to-AI interaction, as your "Agentic AI" proposes Meanwhile, as yet another saying goes... "garbage in, garbage out" - applies to your underpinning of AI data conundrium However.... ....symbiosis requires mutuality - and I am glad you call it out as "future", because one could argue that this isn't the case today! For symbiotic future to be true, those of us that define, design and implement AI, need to consider multiple things: Individual choice (rather than imposition of it) Power balance (large tech today) Ethics (algorithms exploiting human behaviours) The idea of imposition over cooperation is more accurate for many of today’s applications - but let's agree as technologies to ensure we push for a symbiotic future 🤗

Mahadeva Matt Mani

Partner; Global Co-leader Technology & Transformation at PwC

3mo

Sanjay - any thoughts on whether you need to hoard and protect your own data in order to use it well within the enterprise, or do you have to be willing to cleanse and share it with others, so that you can get access to their data? If so, does this further reduce the barriers to competition and differentiation? Also, any discussion about how ensure accuracy of data collection / ingestion at its source? It feels like all the edge agents we have gathering data for the first time will become even more important.

Vyjayanthi Mala (VJ)

Climate Action Enthusiast | Chief, Progress Assessments & Learning | Singer

3mo

Thanks for this top line summary Sanjay Srivastava! My humble submission is calling AI copilots as AI pilots is going to be a double edge sword, as it is still fueling the job-threatening narrative of AI. We should somehow stick to the AI-assistive narrative... let us call the full-fledged "AI copilots" as "AI super copilots" but copilots they should be! Not take the pilot seat!!! :)

To view or add a comment, sign in

More articles by Sanjay Srivastava

  • The Future of Executive Leadership

    The Future of Executive Leadership

    Portfolio Careers and the Connected Leader As I look across the couple hundred global c-level tech executives I work…

    17 Comments
  • The Pilot to Production journey in Generative AI

    The Pilot to Production journey in Generative AI

    Over the Year: Three Lessons Learned We have come a long way with Generative AI. A year ago, I hosted a Stanford…

    11 Comments
  • Synchronizing IT and OT

    Synchronizing IT and OT

    The convergence of OT and IT is gaining traction, driven by the expanding enterprise threat surface and heightened…

    7 Comments
  • What is Digital Transformation?

    What is Digital Transformation?

    Transformation, in the age of digital Digital transformation, as a phrase, can at once be very meaningful and equally…

    8 Comments
  • Geopolitics, the macroeconomy - and enterprise technology

    Geopolitics, the macroeconomy - and enterprise technology

    The role of the CIO is evolving rapidly. Orchestrating and deploying technology solutions is no longer at the top of…

    2 Comments
  • Generative AI in Enterprises

    Generative AI in Enterprises

    Leveraging Generative AI for Enterprise Success Generative AI, a cutting-edge technology that enables machines to…

    13 Comments
  • Innovation at Geographic scale

    Innovation at Geographic scale

    Innovation at Geographic scale – Lessons from India Having lived in and worked across prominent tech innovation centers…

    1 Comment
  • A Personal Reflection

    A Personal Reflection

    The pace of change. Decades ago now, I finished college in India and came to graduate school in the US.

    84 Comments
  • Building Data-driven Businesses

    Building Data-driven Businesses

    Building a data-driven business is clearly on most board agendas. We have all spent time looking at the data…

    6 Comments
  • This blog is written by AI!

    This blog is written by AI!

    As the world continues to digitize at a rapid pace, artificial intelligence (AI) is quickly becoming a critical…

    32 Comments

Insights from the community

Others also viewed

Explore topics