 (8).png)
2 days ago
6
ARTICLE AD BOX
Summary
There’s a tug of war between personalization and privacy. You can guess who’s winning.
One winter night, a year ago, I found myself particularly bothered by a case of freezing feet. Socks wouldn’t help. Moving around didn’t help either - not for long. I decided to test Gemini for a solution. At that time, the chat assistant was a novelty. Gemini takes in the situation and asks me to go make some of my favourite Darjeeling tea and soak my feet in it. Of all the bizarre suggestions…
I thought it was teasing me but couldn’t understand how it knew I loved Darjeeling tea. Then the penny dropped. I had given it what I thought was a set of harmless facts about me — I like learning a language, I sing Hindustani Classical, I’m a tech journalist, and I love Darjeeling tea. So, it tried to personalize its response to me and picked on a tea foot-soak. I laughed it off and used a heating pad —which it could have suggested if it hadn’t been so busy personalizing. I later discovered that tea foot-soaks are actually used somewhere in the world. I went to Settings and deleted the Darjeeling tea.
Personalization Paradox
In just one year, there’s been a big shift in the personalization philosophy of AI companies. Their chatbots don’t wait around for you to make up your mind and give them a few morsels of personal information: they’re always compiling your biography in the background.
I tried a party trick on ChatGPT. I asked it to write me a letter from the future. I left it open-ended. When I read its response, my hair stood on end. ChatGPT seemed to know me better than most of my friends. It had many rather nice things to say and thought I had reason to be much prouder of myself than I was, so I forgave it for being so intrusive.
What felt like a spooky piece of magic back then was actually the data engine at work. In the year since my freezing-feet encounter, AI has transitioned from single-session, forgetful or “stateless” chatbots into "stateful" systems. To cure what the industry calls "operational amnesia" (yes, it’s full of new-fangled and rather bombastic terms) the tedious chore of re-explaining your role, preferences, or writing style every time you log in — AI companies have turned their assistants into permanent behavioral biographers. I can imagine explaining my exact level every time I want a fresh French lesson; listing my weaknesses and strengths and immediate goals. It would not be worth the trouble.
Changed Rules
When we open a sleek, blank text box, our brains naturally treat it like a transient conversation. We drop our guard, assuming our casual thoughts will vanish into the ether when we close the tab. But under the hood, the rules have fundamentally rewritten themselves. Different AI companies approach it differently.
Take OpenAI’s background "dreaming" architecture. It doesn't just wait for you to type out a static list of preferences; it continuously analyzes your past chats, uploaded files, and connected apps to synthesize a fluid, running memory state. Seemingly minor details from months ago are cleverly stitched together.
The system tracks your tone, your late-night anxieties, and your professional shifts to infer highly sensitive attributes about your lifestyle, temperament, and outlook — creating a deeply detailed digital ledger without you ever explicitly hitting Save.
Claude takes another approach and walls off information about you in silos that don’t leak into one another. What you give in your workspace won’t inform, say, your party planning. Google takes a sweeping approach and wants personalization through all of its services — and there are so many of them and we all use them. It can look deeply into your gmail, which is enough fodder for it to know a whole lot about you.
Fighting Back
How AI’s personalization will pan out in the future, is difficult to say. Advancements take place at a breathtaking pace and previously held notions are thrown aside. But if you find AI assistants useful and are uneasy about what it knows about you, there’s no harm developing some AI hygiene in this transition period. Try my trick of asking your go-to AI to write you a personal letter — be specific about what you want. This will tell you what it knows. Go into the settings and rather than being paranoid and deleting everything, choose to keep what helps you and delete what doesn’t.
Use separate chatbots for different tasks. The free tiers give you quite some functionality. By separating them, you’ll be able to manage the information they have better. For sensitive queries involving medical symptoms, financial data, or legal questions, always use the incognito, accountless, or Temporary Chat modes. These ensure your conversation is not saved to your permanent profile, referenced in future chats, or used to train public models.
Also, if your feet are freezing, use a hot water bottle.
The New Normal: The world is at an inflexion point. Artificial intelligence (AI) is set to be as massive a revolution as the Internet has been. The option to just stay away from AI will not be available to most people, as all the tech we use takes the AI route. This column series introduces AI to the non-techie in an easy and relatable way, aiming to demystify and help a user to actually put the technology to good use in everyday life.
Mala Bhargava is most often described as a ‘veteran’ writer who has contributed to several publications in India since 1995. Her domain is personal tech, and she writes to simplify and demystify technology for a non-techie audience.
About the Author
Mala Bhargava
Mala Bhargava was among the first journalists in India to write on personal technology, then known as 'home computing'. With Cyber Media she launched the country's first personal tech magazine, Computers@Home, in 1996. She also wrote a tech trends column, That's IT, for Businessworld magazine for 20 years. She has also written for The Hindu BusinessLine and Fortune. Her speciality has always been writing for 'the rest of us' rather than for the tech-savvy. She has a background in psychology which makes it natural for her to write on how technology impacts everyday life. She is currently a Mint contributor, writing on AI in daily life, specifically the chat assistants. She lives in New Delhi.
English (US) ·