Tamal Chowdhury, a distinguished AI executive and CTO, has an extensive experience in AI-powered product & platform engineering, cognitive automation, data science, machine learning, and next-generation technologies like Big Data, Blockchain and IoT. He specializes in building global innovation structures for critical IP generation, large-scale digital transformation, and commercialization of new ideas to fuel business growth. Currently serving as the Senior Vice President of Artificial Intelligence at Course5, he is a futurist and passionate about leveraging Artificial Intelligence to solve complex challenges of the real-world.
Analytics India Magazine: What are some of the practical implementations of AI that have revolutionized the way humans are functioning?
Tamal Chowdhury: There are seven key areas where Artificial Intelligence has successfully revolutionized the way humans are functioning and is greatly improving our overall quality of life.
- Intelligent Personal Assistants – Digital personal assistants (like Siri or Alexa) have transformed our day-to-day lives to a great extent. We use them for a myriad of tasks like getting information about anything, sending messages and emails, ordering cabs, etc. All these have made our life more comfortable.
- Personalized Services – AI has enabled companies to acquire deep-level understanding about consumers, thus enabling them to serve everyone based on their individual tastes and preferences. This extreme personalization has really propelled our individual customer experiences to the next level.
- Smarter Home Products – AI, in conjunction with the Internet of Things, has made many of our daily-use appliances and products more cognitive. Intelligent sensors to auto-switch off lights, refrigerators to trigger notifications when it detects any issue, home security systems generating alarms, etc. are making our lives easier and more secure.
- Smarter Mobile Applications – AI driven mobile applications like Chat Apps, Translation Services, Cognitive Search, Taxi Services, etc. are enhancing our overall lives and making things easier.
- Autonomous Vehicles – Self-driven vehicles have started bringing about a major transformation in people’s daily commute lives. Fewer accidents, optimal commute routes, ecologically beneficial transportation, etc. are some of the advantages. As AI technologies mature, this is going to get even better.
- Gaming – The gaming industry has historically benefited from AI for more than two decades now. Recent AI advancements, particularly in AR/VR, have taken significantly increased gaming standards and adoption rate.
- Better work quality – Automation of manual, repetitive processes and allocation of human resources to high-end tasks are improving the overall quality of work and life-at-workplaces.
AIM: How is AI affecting your life and work? What are some of the ways that you are personally using AI? Please elaborate on the use cases.
TC: At a personal level, I am a consumer of AI like most people. Smarter mobile applications like translation services, note-taking apps, chatbots, etc. are part of my daily usage. I also use a couple of AI applications that I have developed myself for my personal needs.
In the workplace, I am both a producer and consumer of AI. On the one hand, my team and I build AI products and solutions for clients and our internal teams. On the other hand, we leverage AI to enhance our own quality of work. For example, some of our AI engineering and deployment efforts are themselves driven through machine learning models.
AIM: What are some of the AI-based products or services that you rely on the most? (e.g., chatbot, virtual assistants, phone apps)
TC: Some of the common AI-based products and services that I rely on the most are:
- Smarter Note-taking Apps, specifically Evernote and GoodNotes
- Chatbots like Skype, LinkedIn Chat, Ola and Uber apps, etc.
- My self-developed Spam system
- My self-developed Stock Recommendation system
- Cortana and Google’s speech recognition system
AIM: What are some of the ways that your company is adopting/providing AI services? Please highlight some use cases.
TC: Our company’s core AI strategy is to integrate AI with every strategic business objective and incorporate it in everything we do. We take an AI-first approach.
On the one hand, we are large-scale producers of AI. We develop AI-based products and solutions for the clients and industries that we serve. For instance, our AI Research Suite, Digital Suite, our Discovery program, and many other solutions are learning systems with their features and capabilities being driven by Deep Learning and AI-driven engineering.
On the other hand, we are also consumers of AI. We have leveraged AI to enhance our traditional delivery to clients and to cognitively automate structured and semi-structured processes in our own operations.
AIM: What are the areas of life or employment where you would like AI to be more involved (e.g. medicine, mental health care)?
TC: I would personally like to see AI getting more focus and investments is in solving three major global challenges:
- Healthcare, especially in sub-optimally developed regions like Africa. Even in this 21st century, it is a shame that a significant percentage of global population are still affected by diseases and healthcare issues that are avoidable. Advancements in AI need to be leveraged better to solve many of these problems.
- Global Economic Inequality – The earlier three Industrial Revolutions have generally deepened income inequality in the long run. However, in this fourth Industrial Revolution that we are in the middle of, we have a real chance of addressing this major issue through AI.
- Sustainable Development – Leverage AI-at-scale to better understand the risks of different development options across countries, and choose those options that maximize sustainability.
These areas need strong AI R&D and investments, especially from global companies and developed nations. While there has been an increased focus in the past 4-54 years, much more remains to be done.
AIM: Will AI take away the creative thinking and downgrade the intellectual quotient of humans?
TC: I don’t think this will happen in the foreseeable future. It will be possible only when we achieve Artificial General Intelligence (AGI) that is better than the human intelligence of today. However, AGI is still in its formative years. My sense is that at least two decades of work is needed before AGI can be accomplished with the requisite levels of maturity.
While there will be an improvement in the creative capabilities of AI, surpassing or even matching human general intelligence is still 15-20 years away, if not more.
AIM: What are some of the AI products offered by the company?
TC: Our AI portfolio consists of two major product suites, and client-specific AI solutions.
Our Research AI Suite consists of three key products/platforms – Link, Optimizer Suites and Creative Testing. In this suite, we leverage state-of-the-art Deep Learning techniques (including Computer Vision and Deep NLP) to digitally transform our client industries like market research, advertising, etc.
Our Digital Suite consists of several products and platforms that leverage various AI technologies like Supervised and Unsupervised Learning, and Natural Language Processing, Understanding & Generation to serve our customers. Specifically, our Voice-based Data Discovery platform is a unique example of the global advancements in Speech Recognition and Natural Language Processing.
Our client-specific AI solutions address a wide range of business problems – Image Recognition, Video Analytics, Emotion Detection, Unstructured Text Mining, Cognitive Process Automation, Predictive Modeling, Speech/Audio Recognition, etc.
AIM: How has been the adoption of AI in Indian scenario?
TC: Adoption of AI in the Indian scenario is still at its formative stages. Much of the AI work conducted in India is aimed at global corporations. Having said that, there is a trickle-down effect from global corporations to their India ecosystems. It is also encouraging to note that many niche areas like agriculture and small-scale industries have started opening up to AI, thanks to the Digital India and Start-up India initiatives of the central government, and India-based AI companies.
As a country, we need to focus more on data governance standards, AI-specific policy frameworks, and robust technology infrastructure like high-speed internet networks, mobile and satellite communication, large data centres, etc. Till these things are in place, AI adoption will always remain limited.
AIM: What are some of the challenges that come on the way while adopting AI?
TC: Some of the other challenges to AI adoption are:
- Firstly, there is a genuine lack of global leadership talent in the application of AI. Most of the business leadership chartered with deploying enterprise AI programs are inadequately informed about how AI works, and the idiosyncrasies of AI programs. The problem is compounded by the fact that the technical AI talent is also limited.
- Secondly, many AI technologies still need to go through a maturity process of their own. Till that happens, AI will always get adopted in bits and pieces, particularly by the old economy companies.
- Thirdly, data is still not given adequate attention by many companies. AI thrives on data, and till companies really start investing in Corporate Data Lakes and holistic data strategies, AI adoption will be weak.
- Fourthly, there is a need for a more consistent and standard way of defining AI and its scope. There is a lot of misinformation on AI, its scope, capabilities, how to realise its potential, etc. Different companies interpret AI differently, and some even package non-AI capabilities as AI in a bid to piggyback on the hype surrounding AI. All these make it difficult for companies to objectively assess how AI can help them, and work out practical strategies for the same. This, in turn, makes adoption difficult.
AIM: Will AI be capable of exhibiting emotions and get a conscience? Will that be useful or harmful for us?
TC: Affective Computing is a key aspect of Artificial Intelligence. Emotion AI is arguably its most critical focus area today. There have been some recent innovations in that space, but it is limited in nature. AI systems with human-like emotional behaviour can be built only after Artificial General Intelligence (AGI) is achieved, which I think is at least 15-20 years from today. Also, the scope of emotional behaviour is likely to be limited to a finite set like anger, sadness or happiness.
Generally speaking, AI with emotions will have a positive impact. It will ensure a more efficient HCI (Human-Computer Interaction), thereby enhancing the value of AI agents.
Note that AGI won’t be enough to develop Conscience in AI agents. That would need the most advanced version of AI known to man today – Artificial Super Intelligence (ASI). However, ASI is largely in the realm of science fiction as of today.
Try deep learning using MATLAB