Sunday, December 13, 2015

OpenAI: AI for All

Yesterday saw the announcement about the birth of OpenAI, a non-profit organization to develop and advance Artificial Intelligence (AI) technologies, and share these in the greater good.  Infosys, and I, are a part of this endeavor, and very excited about it and I've been asked tons of questions in the last 24 hours about this, so I thought I'll write some thoughts down.


A few weeks ago, Marvin Minsky, one of the fathers of Artificial Intelligence (AI), the one who gave the field its first definition -- that AI is "the science of making machines do things that would require intelligence if done by men" -- made a few sobering statements about the state of the art in AI.  Indeed I felt sad listening to this giant lament the lack of fundamental progress in the field, and highlight some of the underlying causes.  And this despite all the buzz and hype AI work has picked up lately.  Marvin is one of the truly great human beings and scientists, whose teachings and advice helped influence my life and led me to focus on AI in my grad studies and beyond.  So more than anything else, I see OpenAI as an opportunity to "do something about it".

My friend and teacher Alan Kay once referred to Sam Altman as a "builder of civilizations".  When Sam, a wise man who is but 30 years old, was thinking about the idea of building an open ecosystem for, among other endeavors, AI, Alan and I shared our ideas with him and our experiences.  Sam asked me if I would be ok with the fact that such an endeavor would be untethered and would produce results generally in the greater interests of humanity, and he was somewhat surprised by my reaction, that indeed I would only support this venture if such an openness was a fundamental requirement!  In all my experience with corporate research teams, I found a continual struggle for the teams to find relevance with the work in the "here and now", usually knowing that this unnecessary and premature seeking of relevance not only blinds us to those opportunities that can shift our paradigms, it defeats the point of research.  He shares the view that cooperation helps dramatically improve our lot, helps create a foundation for much larger value creation than any isolated "feudal" system can.  Indeed, endeavors such as agriculture, and science, show that when we share, we improve all of us.  As Newton once said, "if I have seen further, it is by standing on the shoulders of giants".

I am really excited that Sam, Elon Musk, and others -- including Reid Hoffman, Peter Thiel, and Amazon Web Services, and Infosys of course -- are supporting this great endeavor, and in addition to Alan, a great set of leaders will serve as advisors.  Ilya Sutskever, who has worked over the last several years as a leader in developing the so called "deep learning" techniques, will direct the research at OpenAI, and will be joined by several dynamic leaders and individual contributors, whose conviction and imagination will help this field move forward, based on the best of what we know, the best ideas, the best inventions, and the key lessons.

Our wish is that together the OpenAI team will do unfettered research in the most important, most relevant dimensions of AI, no matter how long it takes to get there, not limited to just identifying dancing cats in videos, but to creating ideas and inventions that amplify our humanity, that help us learn more, see/perceive and understand more, and be more.

Why Open?

One question that's been asked since yesterday, is why should this be open?  Isn't it better to have deep AI be in the hands of a select few experts or specialists?  My sense is, our trust in complex systems stems mostly from understanding these and their predictability, whether it is nuclear reactors, lathe machines, or 18-wheelers; or of course, AI.  If complex systems are not open, not open to be used, extended, and learned about, they end up becoming yet another mysterious thing for us, ones that we end up praying to and mythifying.  The more open we make AI, the better.

Why Infosys?

Another question that's been asked a lot, is why Infosys?  We at Infosys, with over 150k software engineers, are unique beneficiaries of and contributors to this endeavor.  Most of our work is in building and maintaining software systems, and AI will increasingly shape the construction and evolution of intelligent software systems, in all kinds of domains and industries.  In addition, as a large services company, many parts of our work can transform fundamentally with AI.  In services like infrastructure management, business process outsourcing, and verification and maintenance of existing software, we can massively migrate mechanizable work to automation, and instead build intelligent software systems, that amplify us, our abilities, as well as those of our customers.  So a great transformation that we are undertaking at Infosys, is to embrace automation at a very large scale, so people can, as Prof Mashelkar once said, "do more with less for more", and at the same time, educate ourselves in new areas to help build intelligent systems, but also to innovate in our work, to exercise our creativity in everything we do, and amplify our abilities, our humanity, using AI.

But beyond business, there is another key reason; our endeavor to do purposeful work.  Our founders always believed in this.  Many years ago, Mr Murthy and our founders started the ACM Infosys award, which celebrates great young Computer Science practitioners.  The Infosys Science Foundation supports work in the pure sciences, as does the Infosys Foundation in India.  And most recently, the Infosys Foundation in the US, is working hard on its mission to help enable/expand computer science education, and has already started many promising initiatives here in the US.  So OpenAI aligns very nicely with our long-held values.

So as we get started on this great journey, I find myself excited, hopeful that OpenAI will help uncover great innovations, that new AI techniques yet to be discovered, and built to share, built in ways that are open to all, will help us transcend our limitations, improve and amplify us all, and that our work in artificial intelligences, may help bring us closer to our natural spirituality...

Palo Alto,
Dec 12, 2015

Saturday, August 1, 2015

Days, Quarters, Years. The Moments in Our Momentous Lives...

Today marks an year since I started my journey as the CEO of Infosys.  One Year.  One spin of the earth around the sun, with my life's work centered around the Infosys planet.  There has been a ton of interest about my first year.  This, of course, is a good thing.  I feel privileged to be in this position, and feel ever more aware of the weight of the responsibility.  But I don't understand the significance of the one year mark.  I never did.  Anniversaries and artificial ritualized celebrations of the sort, beyond the uniquely personal aspects of these, somehow always seemed to me to be pointless constructions.

We posted a good quarter recently.  Our Q1 results saw the best revenue growth in 15 quarters.  Most of this was due to great execution by our sharply focused leadership teams, and some of it due to the innovation seeds that we've sown over various points last year, that are starting to bear fruit.  But again, the 90-day cycle, an imposition largely constructed by public markets around the world, seems to defy any meaningful purpose.  Indeed, arguably, focus on 90-day performance, can often distract us from longer-term progress.  When we look at many emerging and flaring crises in the broader economy around us, we can see the results of some of the short-term gain oriented decisions and mindsets.  Whether in the debt matter between the Eurozone and Greece, or Italy, or the stock market situation in China or even many industries' responses to the disruptions they face, etc...  We all see large-scale phenomena over the centuries that are collaborative in nature, whether agriculture or science or long-term research or some forms of democracies, and the fruits they continually yield, and yet we, by and large, make like hunters and gatherers and force ourselves into local minimas, knowing that the right things to do have far longer cycles than 90 days.  There are far too few examples of long-term innovation, and waiting for fruits to be borne over longer lifecycles than our attentions and our senses can stay tuned for.

And yet we too are governed by our cycles.  Our life cycles.  Earlier this week Prof. Kalam passed away. An extraordinary, and extraordinarily humble and grounded human being, one that set an almost incomparable example of what a human can do, and at the same time showed us that a human can do so if we get ourselves to.  While doing something he loved, his amazing life came to an abrupt end.  Once again, a reminder of the fragility, and yet the finality of our lives.  We shine, sometimes brightly, for a while, usually an all too brief a while, and then we fade away, into the same grand void the same grand dance that creates us.

So all this got me thinking this weekend, on our moments, our anniversaries and our lives.

I was in Munich yesterday and the day before, visiting some clients and meeting our team.  Had some amazing sessions and discussions.  But the visit with our onsite team was what stayed with me.  Young kids and experienced Infoscions.  Full of passion and energy.  Inspired.  Of course there were the selfies.  But it was the glint, the gleam, in their eyes that I found inspiring, and awesome.  They were all bringing innovations to their day-to-day work, and were excited about sharing these.  I asked them about their lives in Munich, the long summers, the harsh winters, the language, the food.  Most of the colleagues were young, a lot of them singles, yet to marry.  Living away from families and loved ones.  One young male infoscion said he's had to learn to cook.  That there is a local grocery store (called Bollywood!).  But their energy, the human energy, was palpable.  And on my way back to the hotel, I thought this was what it was all about; this urge to innovate, to do more, to be more, to do something beyond us, to improve the lot around us, every day, every single day.  To wake up every day and to work to improve things.  To deal with the struggles and stresses of our lives, and yet to work to endure and to improve.  To renew all we are and all we do, and yet bring some new in addition.  To marry our natural daily cycles, with longer-term improvements and patience.  Living in the moment and yet not being blinded by instant gratification.  So the most heartening bit of all was to see this basic duality at work in our teams, and that lifted me up.  I found that to be the most fitting realization for today.  And if today is anything like yesterday was, anniversary or not, it will have been a worthwhile day, a day to remember for all the right reasons, a great piece of our extraordinarily fragile, immortal, lives.

Saturday, January 24, 2015

Davos, a state of mind

I wrote the following thoughts on our internal Infosys portal, and felt it is worth sharing with the wider world...


Davos is a small town in Switzerland, but, for a few days in January, during the World Economic Forum’s gathering, it transforms into a large state of mind.  I’ve been here with our amazing, inspired, Infosys team, meeting with clients, partners, world leaders, thought leaders and other leaders and visionaries.  Beyond the learnings and the dealings, there are the journeys Davos puts you on.  It is easy to get lost here, in more ways than one.  In the maze of routes and security gates that connect the various venues.  But also in the plethora of thoughts, visions, roadmaps, forecasts, warnings and other articulations, for these are to be found all around, aplenty.  I’ve been absorbing much of this, over the last 3 days, and every once in a while have also been adding to the mix, answering the questions of curious journalists, and in other venues.  It is somewhat surreal; there is a lot of human warmth in the freezing alpine cold, a very curious and spontaneous blend of diversity, of thought, of personalities, of ego, and of humanity in general.  The atmosphere, in many ways, drowns out the content.  And all you remember, to paraphrase Maya Angelou, is how you felt.  As I look back on the last 3 days, I feel good, about the future of the world, the future of our company.  For sure much doomsday prognostication was to be found, and real threats analyzed, but by and large, as I get ready to visit a great client and partner of ours, at 11PM, I find myself with a distinct feeling of comfort, that the road ahead of us is ours to carve, ours to shape, that while there are great big factors and forces that can influence our world and forces that can shape our contexts, that the stage has been set for us to do our thing, and that if we do it, the greatness to follow is ours to achieve.  That the times ahead are calling on us to be bold, to be decisive, to be determined, not to sway in the winds driven by others, but in building the great futures that our clients seek and that we deserve.  So in that sense Davos has been a great mirror, a great echo chamber, one where you travel far to reach, and get lost in the brilliance of humanity, only to find that which you always knew, but perhaps had lost the courage to believe…

I leave here as a world citizen, a proud Infoscion, ready to, together with you, all of us together, to build the future of our predictions, the future of our aspirations, a more human future.  And that, underneath the layers, of snow and of words, is what I've found this town to be all about.