Somewhere between \u201cplease walk me through your experience\u201d and \u201cdo you have any questions for us,\u201d a new request has entered the group chat of modern hiring.<\/p>\n\n\n\n
\u201cShare your ChatGPT history.\u201d<\/p>\n\n\n\n
That is not a punchline. Apparently, it was a hiring decision.<\/p>\n\n\n\n
A founder on LinkedIn claimed he was unsure about a candidate after their conversation, so he asked for the candidate\u2019s ChatGPT browsing history. He liked what he saw. The candidate got hired. And the internet did what it does best… turned it into a debate, a warning, a flex, and a future prediction, all at once.<\/p>\n\n\n\n
Our WhatsApp chat had the full range, from immediate suspicion to existential dread to \u201cokay but what if it\u2019s actually useful.\u201d<\/p>\n\n\n\n
Kiranmayi said what many of us thought first. It\u2019s a weird ask, and it also smells like the kind of LinkedIn post designed to travel, not to tell the truth. Inderpreet called it \u201cuniversal stupidity\u201d.<\/p>\n\n\n\n
Because even if it happened exactly as described, it raises a bigger question.<\/p>\n\n\n\n
What <\/strong>kind of workplace do we create when private logs become proof of competence?<\/em><\/p>\n\n\n\n On paper, the founder\u2019s logic sounds simple. \u201cI couldn\u2019t read this person in conversation. So I\u2019ll read how they think when they\u2019re alone.\u201d<\/p>\n\n\n\n Bhavana offered the most coherent version of that argument. She treats ChatGPT like a junior team member, and her prompts are detailed briefs. In that framing, a candidate\u2019s prompts could reveal how they structure thinking, how deep they go, and how they evaluate outputs. Tina echoed a similar nuance. If the role requires research, analysis, forecasting, or written depth more than verbal articulation, a peek into how someone explores ideas could feel relevant, provided it is voluntary.<\/p>\n\n\n\n Jaideep added a reality check. Prompting habits might show initiative and learning style, but there\u2019s no strong proof that people are being hired solely on history, and over-reliance on AI often shows up as a negative in generic applications or live interviews.<\/p>\n\n\n\n So yes, there is a legitimate question hiding here. Can AI use be a professional skill signal?<\/em><\/p>\n\n\n\n But here\u2019s the catch.<\/p>\n\n\n\n A recruiter asking for your ChatGPT history is not asking for a portfolio. They\u2019re asking for access. That is certainly not an assessment.<\/p>\n\n\n\n Sreeparna\u2019s point, ChatGPT doesn\u2019t only contain job-related searches. It can contain anything private. That is exactly what makes it useful to humans in the first place. People ask AI the questions they don\u2019t want to ask anyone else.<\/p>\n\n\n\n Puspanjalee\u2019s and Harshita’s examples are funny because they’re true. Your history might be \u201chow to fix a bulb\u201d, \u201cwill Boroline work on my elbow?\u201d, “embroidery colour suggestions…” None of this is \u201cwork.\u201d It\u2019s just life.<\/p>\n\n\n\n Reubenna called the trend unacceptable because it highlights how unsecured our lives already are. Janaki said she\u2019s now going to be careful what she uses ChatGPT for, which is honestly the saddest outcome here. Not because you lose privacy in the moment, but because you start policing your curiosity pre-emptively.<\/p>\n\n\n\n Nibha raised an even more practical point. Many people use enterprise accounts where history can\u2019t be shared outside the organization. Even if someone wanted to comply, they literally can\u2019t. And they shouldn\u2019t have to.<\/p>\n\n\n\n This is the line that matters. Public social media screening is not the same thing as private AI logs.<\/em><\/p>\n\n\n\n Ruchi asked, \u201cwhat is privacy anyway,\u201d pointing out that social profiles get monitored for visas and sometimes screened by organizations. Disha\u2019s reply is the boundary we should not blur. Those checks rely on publicly available information, not private search history. Public behaviour is one thing. Private logs are another. We cannot normalize a workplace that has no boundary between employee and person.<\/p>\n\n\n\n Inderpreet\u2019s point made us think. The founder shouldn\u2019t have asked, the candidate should have refused, but maybe they needed the job more than privacy.<\/p>\n\n\n\n That\u2019s the heart of it.<\/p>\n\n\n\n In hiring, \u201cconsent\u201d often comes with an invisible footnote. If you say no, it may cost you.<\/em> Even if nobody says it out loud.<\/p>\n\n\n\n So yes, you can argue \u201cif the candidate is okay with it, why object?\u201d But that logic collapses when you remember how power works. Interviews are not equal ground. They\u2019re not a friendly exchange of boundaries. The moment private history becomes a filter, people will comply to survive.<\/p>\n\n\n\n Even if we ignore the ethics, there\u2019s a basic reliability problem.<\/p>\n\n\n\n You can manipulate interactions, curate a \u201chireable\u201d history, and optimize your prompts the way people optimize resumes. Once something becomes a hiring signal, people learn to game it. Quickly.<\/p>\n\n\n\n Pradeep also pointed out he doesn\u2019t understand how chatbot history can reliably reveal job skills. A history can show curiosity, sure. It can also show randomness, stress, health anxiety, relationship questions, late-night thoughts, and things that are nobody\u2019s business.<\/p>\n\n\n\n So the founder might think they\u2019re reducing margin of error, but they might simply be swapping one kind of uncertainty for a messier one, with more harm attached.<\/p>\n\n\n\n If the interviewer needs AI to help decide, their cognitive skills should be questioned. Felicia added a line that deserves to be printed on every interview training deck! An interviewer\u2019s job is to extract competence through dialogue. If they can\u2019t tell you\u2019re smart unless they read your private logs, their ability needs questioning.<\/p>\n\n\n\n Interviews matter because people can surprise you. Resumes don\u2019t capture everything. AI suggestions don\u2019t either. If you want to assess management skills, you put a person in a situation and see what they do. You don\u2019t need a bot to tell you who they are. Kavitha summed up the whole dilemma in one line. AI is a good servant but a bad master.<\/p>\n\n\n\n If the role is actually about prompt engineering or heavy AI use, then test it. Live. With a scenario aligned to the job description. Asking for history is invasive and unreliable.<\/p>\n\n\n\n Nibha echoed, add a round to check it on-call. Give a task. Observe thinking. Don\u2019t demand diaries.<\/p>\n\n\n\n Also, if someone doesn\u2019t use ChatGPT for work, they shouldn\u2019t be penalized. Not using a tool is not a lack of intelligence. It\u2019s a preference, a workflow, sometimes even an ethical choice.<\/p>\n\n\n\n If you want to see how someone briefs, structures, and evaluates, you can do that without peeking into their personal questions.<\/p>\n\n\n\n Sherein compared it to the classic \u201cscience: boon or bane\u201d debate. It does feel like that, but with a 2026 upgrade. The boon isn\u2019t just efficiency, it\u2019s access to knowledge. The bane isn\u2019t just dependency, it\u2019s surveillance.<\/p>\n\n\n\n We already live in a world where we can\u2019t fully trust what we see or hear. Managers doing this only deepens the lack of trust. <\/p>\n\n\n\n This is the crossroads. Do we respond to distrust by building better conversations, or by building bigger audits?<\/p>\n\n\n\n Because once we decide private logs are fair game, we don\u2019t just change hiring. We change what people feel safe asking, learning, and exploring.<\/p>\n\n\n\n AI can absolutely supplement human intuition. It can help reduce errors, speed up research, and improve quality of work when used thoughtfully.<\/p>\n\n\n\n But a candidate\u2019s ChatGPT history is not a fair screening tool. It\u2019s a private space where people think out loud, ask small questions, ask scared questions, ask silly questions, and sometimes ask the questions they don\u2019t want attached to their name.<\/p>\n\n\n\n If a workplace needs access to that to feel confident about you, it\u2019s not really hiring. It\u2019s profiling.<\/p>\n\n\n\n And if we normalize that, the cost won\u2019t just be privacy.<\/p>\n\n\n\n It\u2019ll be curiosity.<\/p>\n\n\n\n Did this conversation make you think?<\/p>\n\n\n\nWhat recruiters think they\u2019re getting vs what they\u2019re actually asking for<\/h3>\n\n\n\n
Why this feels like a privacy violation…because it is one<\/h3>\n\n\n\n
\u201cBut the candidate agreed\u201d is not a moral pass<\/h3>\n\n\n\n
The other uncomfortable truth. This is also a terrible metric<\/h3>\n\n\n\n
If a recruiter can\u2019t assess competence in conversation, who needs the help?<\/h3>\n\n\n\n
So what\u2019s the sane middle?<\/h3>\n\n\n\n
The real question we\u2019re avoiding<\/h3>\n\n\n\n
What do we think?<\/h3>\n\n\n\n