What are you going to do in an economy when there is going to be massive displacement due to AI? Can AI chatbots replace teachers? Do you have to worry about your brain being scanned for targeted ads? These were some of the questions that were addressed during the annual Synapse Conclave in the capital last month, where over 40 speakers, including Nobel laureates, Pulitzer Prize-winning journalists, neuroscientists and philosophers, joined AI pioneers to discuss what it means to be human in an age of thinking machines.
“The core intention was the intersection between science, tech and society,” says Shoma Chaudhury, journalist and founder-director of Synapse and the firm behind it, Lucid Lines Productions. While last year’s edition matched AI with quantum technology and biotechnologies, this year the team focused on the brain, life sciences and bio engineering. “AI and the life sciences — material sciences, climate change — they are all converging. Typically, science and tech conferences are very B2B. For instance, you will have a crypto conference or fintech conference or an AI conference. Here, we were making visible the intersection between all these sciences, its potencies and opportunities but also its ethics and impacts on society,” Chaudhury continues. For the 1,500-strong audience, which included entrepreneurs, doctors, 10 sitting High Court judges, Congress chief Rahul Gandhi, people from the ministries, architects, artists, even those from Bollywood, there was a sense of learning something entirely new and its relevance to one’s life.
Lawyer and AI ethicist Nita Farahany at the Synapse Conclave in New Delhi. | Photo Credit: x.com/nitafarahany
Take, for instance, the debate around technology giving access to the inner workings of the brain and the need for laws to regulate this. Lawyer and AI ethicist Nita Farahany, a Synapse speaker also known for her TED talk on the right to mental privacy in the age of AI, has often highlighted how we are quickly moving into a world where what one is thinking and feeling is just as transparent and can be just as easily decoded using AI and neurotechnology. Last year, in an interview with Manoush Zomorodi, the host of TED Radio Hour from NPR, she pointed out that while “consumer neurotech devices could finally enable us to treat our brain health and wellness as seriously as we treat the rest of our physical well-being and that regular use of brain sensors could even enable us to detect the earliest stages of the most aggressive forms of brain tumours... all of this will only be possible if people can confidently share their brain data without fear that it will be misused against them”.
These concerns were reiterated by Carme Artigas when I spoke with her at Synapse 2025. Artigas has served as a co-chair of the UN’s AI Advisory Body from 2023 to 2024 and was also secretary of state for digitisation and AI in Spain. As an AI safety advocate, these days she appears as much an inhabitant of an uncertain frontier as the scientists, considering what her domain has been up against (AI safety has gone out of style, what with the U.S. and the U.K. refusing to sign on to a joint statement at Paris’ AI Action Summit emphasising safe and responsible use of the technology).
“There’s no one in the world who disagrees with the need to develop AI in an ethical way,” Artigas says. “But companies use this as an argument to say they don’t need regulation because they already have their code of ethics. Well, ethics is subjective, and changes from country to country and from CEO to CEO.” Artigas points to an EU ban on scanning brainwaves. “Immersive technology already exists, like in hospitals. The problem is when this goes out of the hospital and it becomes a consumer good… if I buy earphones from Apple, Apple has a patent already registered where they can read your mind through your earphones. Then what is the ethical use of that? Who is preventing a commercial company from maximising their benefit by using your brainwave data and selling it to a consumer?” As things stand, outside the European Union, there’s largely more regulation of household appliances than there is of AI.
(As told to Poulomi Chatterjee)
These were the conversations that took centrestage at the conclave, as everyone from adman Prahlad Kakkar to filmmaker R. Balki from the audience paid attention to the speakers. While the event opened with robot Sophia — which rolled onto the stage, answered some light questions and had the crowd whipping out their phones — attendees now familiar with generative AI were more curious about what happens when we take these capabilities and plug our brains into them (or plug them into our brains)?
“People essentially want to live longer and healthier lives,” admits Elena Sergeeva, an advisor at the Foresight Institute, which isn’t just looking at brain-computer interfaces but is offering grants to leverage such brain-computer links to compete with the super-intelligent AIs of the future. Just like some of the other speakers, her work also looks into using biotech to extend lifespans. While the marriage of AI and biotech may seem new, Sergeeva points out that these are technologies that have long been intertwined. AlphaFold, first released in 2018 by Google parent Alphabet, has been used in predicting protein structures, a critical process that is of use to pharmaceutical researchers.
“AI has already been helpful in drug discovery,” Sergeeva says. “In a few years, it will not be possible to do biotech without any AI support.”
Elena Sergeeva, neuroscientist inventing new bio-forms, at Synapse Conclave. | Photo Credit: (Courtesy Synapse 2025)
So what if these advances — like arresting ageing — burden the world with overpopulation, or what if they remain available only to the rich? “Humans are ‘transhumanistic’ all the time,” Sergeeva responds. “In the past, nobody could imagine that if you have bad vision, you can just use glasses… Technologists will develop in synchrony. When technologies appear, they are kind of a luxury, but over time, they become available to everyone. Businesses compete to make it cheaper and better.”
Synapse founder Chaudhury says these conversations reflect how young tech entrepreneurs are thinking in the country. Awais Ahmed, founder and CEO of Bengaluru-based space-tech start-up Pixxel, is a prime example, she says. “Extremely tech-savvy and optimistic, they believe in techno-solutionism. It has opened my mind, even when I debate or question some of their enterprises. They don’t think death, age, or the nature of human intelligence is inevitable. They believe that intelligence itself can evolve into something else. It is disturbing at some levels and eye-opening at another,” she says.
(As told to Poulomi Chatterjee)
Earlier that day, when I opened Instagram, the platform had an ad featuring someone selling shared GenAI subscriptions on the cheap. The last time I saw such illegal ads on the social app was for shared Netflix subscriptions. More than 65% of Indians polled by Microsoft said they used AI, over double the global average of 31%. And Indians are using AI more and more for a host of things: it’s not just about automating homework or writing polite emails with formal English. An increasing number of young Indians are using services like Character.AI to substitute real world friendships, or even emulate a relationship. This is particularly interesting for the consequences it can wreak on young people resorting to AI for guidance at sensitive times in their lives (see box).
Beyond making digital friends and romantic partners, though, there’s ‘vibe coding,’ a term invented by the computer scientist Andrej Karpathy. Give a chatbot information on what you want a piece of code to do, and it handholds a user on how to deploy it. I have no coding background, and while writing this paragraph, I was able to write a piece of code on my phone that will now alert me whenever the Union government’s gazette website puts out a notification for the ministries I cover. This may well take off massively among young IT graduates, with many tinkering with unfamiliar coding environments, and perhaps successfully fooling prospective firms about their actual capabilities. Or they could just make really cool code for themselves.
Then there are agentic models, which can take over an entire device and automate tasks end-to-end, as though a human was going through the steps. At the very least, we can expect something like “Send good morning messages to each family member on WhatsApp”. At the limit, imagine autonomous embodied AI bots that, unlike Sophia, can actually take chores off your hands. On the industrial side, Automation Anywhere, a San Jose-based firm that develops AI code for robotic automations, said 40% of its deployments globally were in India.
Filmmaker Vikramaditya Motwane (left) at a panel discussion at Synapse. | Photo Credit: (Courtesy Synapse 2025)
So, should one worry about AI taking over a big chunk of our lives as we know it? “The thing is AI creeps up on you bit by bit,” says another Synapse speaker, director Vikramaditya Motwane (of Udaan and CTRL fame). AI tools are already being used in music and visual effects, he adds, and there may soon be specialised software to automate writing processes. “As long as this technology is as expensive as making a film in real time, people won’t bother. The moment the technology catches up, AI is going to be at half price versus us at full price. Then what’s the industry going to do?”
My use of AI in my blog Masala Lab is significantly high in its impact. I have built an AI agent that is a personalised tutor. So, every couple of hours, it pops up on my system and gives a brief course on the subjects I wish to learn. There’s another AI agent that’s trained on everything I’ve written, my notes and my book, so it’s able to generate the first draft of a script for my YouTube videos. I then tinker with it and improve it. But the AI is 99% better than any intern I could have hired. I have built these tools using open-source AI chatbots from ChatGPT, Llama, Gemini, Perplexity and Claude.
There’s also a great AI podcast tool called NoteBookLM from Google that reads out articles I missed out on the previous day. So, when I am driving, I listen to them and can even interrupt the host to ask questions of my own. I have also built a personalised tutor for my child, implementing specific guardrails around sensitive topics. The AI assistant has been optimised to use language suitable for a 12-year-old. I have finetuned it so it doesn’t give out answers immediately and gives my son a moment to really think. I am also signed into my account so I can check the history and know what questions he’s asked. AI can basically augment what a parent does while also keeping in mind the cultural values we have
Krish Ashok, food science influencer, techie and author of Masala Lab
****
At work, I use ChatGPT, DeepSeek and Grok primarily for research when I am looking into new markets, competition, or companies. I also use them to rewrite drafts for social media posts, emails and investment memos. Of these, DeepSeek is the best for any reasoning problems and research, ChatGPT and Claude are good for storytelling, and Midjourney or Grok are great at generating images.
Among the new, cool AI tools, Apple’s AI generated transcript for podcasts and readalong is amazing and very accurate; the new note-taking capability from Gemini on Google Meet is phenomenally helpful during meetings, and if you’re a developer or want to get started, Bolt.new is excellent for building and deploying simple apps quickly.
Gowri Shankar Nagarajan, associate partner, Antler India
(As told to Poulomi Chatterjee)
Published - March 07, 2025 06:09 pm IST
