LLexfolio

Tag: CERT-In

  • Cyber Crime Scenarios in Offices and Colleges in India: 6 Situations

    Last updated: 12 November 2026

    Rules read easily until a Monday morning when the accounts head says a vendor’s bank details changed “last week” and Rs 8 lakh has already gone. This guide walks through six realistic situations that offices, colleges and schools in India face, and for each one sets out the legal position and the sensible first moves. All the people and facts below are invented illustrations; they are not real cases.

    The laws used throughout are the Information Technology Act, 2000 (IT Act), the Bharatiya Nyaya Sanhita, 2023 (BNS), the Bharatiya Nagarik Suraksha Sanhita, 2023 (BNSS) and the Bharatiya Sakshya Adhiniyam, 2023 (BSA). Outcomes depend on evidence and on what a court finds, so nothing here predicts how a case will end.

    Learn this properly. Lexfolio offers Cyber Crime Law for Offices and Institutions, a self-paced certificate course on cyber crime law for offices and institutions. Course price: Rs 599.

    Illustration 1: The vendor whose bank details “changed”

    Facts (illustration): A trading firm receives an e-mail that looks like it is from a regular supplier, saying its bank account has changed. The accounts officer pays an invoice of Rs 8 lakh to the new account. The real supplier calls three days later asking about payment.

    Legal position. A person who dishonestly induces delivery of money by deceiving another commits cheating (section 318 BNS), and pretending to be someone else is cheating by personation (section 319 BNS, punishment up to five years or fine or both). Using a computer resource for the personation also attracts section 66D of the IT Act (up to three years and fine up to Rs 1 lakh). Because the firm’s own officer approved the payment, the Reserve Bank of India’s 2017 circular on *unauthorised* electronic transactions is unlikely to give a refund right.

    What to do. Call the bank immediately and ask it to contact the receiving bank to hold the funds, then confirm in writing. Report on cybercrime.gov.in or call 1930 on the same day, and file a police complaint with the e-mail in full header form. A hacked mail account is also a CERT-In listed incident (unauthorised access, phishing), so report within six hours of noticing. Afterwards add a rule that bank detail changes are confirmed by a call to a known number.

    Illustration 2: The departing manager and the client list

    Facts (illustration): A sales manager resigns on a Friday. Logs show that on Thursday night he forwarded the full client database and price lists to a personal e-mail account. On Monday a competitor contacts three of the clients.

    Legal position. Copying data from a computer without the permission of its owner is covered by section 43 of the IT Act, which makes the person liable to pay compensation, and by section 66 where it is done dishonestly or fraudulently (up to three years or fine up to Rs 5 lakh, or both). Disclosure of personal information obtained under a lawful contract, with intent to cause wrongful loss or gain, is separately punishable under section 72A (up to three years or fine up to Rs 5 lakh, or both). The employment contract’s confidentiality clause and any civil remedy for misuse of confidential information sit alongside these. Whether the manager acted dishonestly is a matter of proof, and a court will look at the contract and whether the data was within his authorised access.

    What to do. Do not alter his mailbox or laptop. Image the laptop and keep the mail server logs, which the CERT-In directions already expect covered entities to hold for 180 days. Send a written notice to preserve and return the data, and take legal advice before deciding between a police complaint and a civil injunction. Check whether the client list contains personal data, because that will matter under the data protection regime from 13 May 2027.

    Illustration 3: The fake admission website

    Facts (illustration): A private college learns that a website with a near identical name and its logo is collecting “admission confirmation fees” through a UPI QR code. Parents have started calling the office.

    Legal position. Collecting money by pretending to be the college is cheating by personation and, because it is done through a website, an offence under section 66D of the IT Act. Using the college’s name and logo may also infringe its trade mark and copyright, which are civil and criminal matters under their own Acts. The fraud affects the public, so quick action matters more than a perfect case.

    What to do. Issue a clear public notice on your real website and notice board giving the genuine payment details. Report the fake site to the domain registrar and hosting provider with proof of your rights, and complain on cybercrime.gov.in. Tell affected parents to call 1930 and their banks immediately. Keep dated screenshots with the full web address visible. If a college runs its own servers, check whether the site is hosted on, or links to, anything of yours.

    Illustration 4: The morphed image in a class group

    Facts (illustration): A student circulates a morphed, sexually suggestive image of a classmate in a messaging group. The affected student is a minor.

    Legal position. Publishing or sharing private images or sexually explicit material electronically engages sections 66E, 67 and 67A of the IT Act, and where the material shows a child in a sexually explicit act, section 67B (up to five years on a first conviction). Offences involving a child are also dealt with under the Protection of Children from Sexual Offences Act, 2012. The seriousness is much higher where a minor is the subject, even if the sender is also a minor, and the juvenile justice framework will apply to a child accused. An institution should not try to decide such a matter internally.

    What to do. Support the affected child first and inform the parents. Do not forward, print or keep copies of the image yourself, even as “evidence”; secure the device and hand it to the police. Report to the police and the cyber crime portal and follow the school’s child protection procedure. Ask the platform to remove the content through its grievance process. Institutions should know their duties under the POCSO framework before an incident, not during one.

    Illustration 5: The ransomware message on a Sunday

    Facts (illustration): A diagnostic clinic’s reception computers show a message that all files are encrypted and a payment in cryptocurrency is demanded. Patient records are stored on the same server.

    Legal position. Introducing a virus or ransomware into a computer system without permission falls under section 43 of the IT Act and, if dishonest, section 66. Ransomware and data breaches are listed incidents under the CERT-In directions, so the clinic must report within six hours of noticing. The directions say failure to comply can invite action under section 70B(7), which provides imprisonment up to one year or a fine up to Rs 1 lakh, or both. Patient data is sensitive personal data, so section 43A (compensation for negligent security practices) is relevant today, and the DPDP Act breach intimation duties will apply from 13 May 2027.

    What to do. Isolate the server from the network but do not wipe it. Report to CERT-In and the police, note the time of discovery, check whether back ups are intact and offline, and take forensic help before restoring. Paying a ransom guarantees nothing, and the police should be told before any payment is even considered.

    Illustration 6: The “principal” asking for urgent transfers

    Facts (illustration): The finance clerk of a trust receives a WhatsApp message from an unknown number, with the managing trustee’s photograph, asking for Rs 2 lakh to be sent to “a supplier” and a few gift vouchers to be bought.

    Legal position. Pretending to be a known person to obtain money is cheating by personation (section 319 BNS) and section 66D of the IT Act where a computer resource is used. If the clerk transfers the money, it was an authorised transfer from the clerk’s account side, so recovery rests on the speed of the report and the bank’s ability to hold funds.

    What to do. Do not pay. Verify any instruction to pay by calling the person on a number you already hold. Screenshot the profile, number and chat, report the number on WhatsApp, and file a complaint on cybercrime.gov.in. If payment was made, call 1930 at once. Add a written rule that no payment is made on a messaging app request alone.

    Common thread

    In all six situations, three habits decide the outcome: report fast, preserve the original evidence, and know in advance who in the organisation will do what. Where personal data is involved, the legal duties will tighten from 13 May 2027 when the main DPDP obligations are scheduled to apply.

    Learn this properly. Lexfolio offers Cyber Crime Law for Offices and Institutions, a self-paced certificate course on cyber crime law for offices and institutions. Course price: Rs 599.

    Frequently asked questions

    Can an institution be held responsible for what a student or employee does online?

    A: The person who commits the offence is primarily responsible. An institution can face separate consequences if it ignored a reporting duty, failed to keep reasonable security or has specific child protection duties. The answer depends on the facts and the law that applies.

    Should we file an FIR or only a portal complaint?

    A: File the portal complaint for speed, and also follow up with the police station for a formal FIR. For financial fraud call 1930 straight away so the bank can be alerted.

    Do we need to tell the person whose data was stolen?

    A: From 13 May 2027 the DPDP Rules require a data fiduciary to inform each affected person without delay and the Data Protection Board with a detailed report within 72 hours. Before then, there is no single general rule, but sector regulators and contracts may require it, so check yours.

    Is a WhatsApp chat admissible as evidence?

    A: It can be, if proved as an electronic record under section 63 of the BSA with the required certificate. Keep the original device and the chat export, and do not delete anything.

    Can we trace the person behind a fake account ourselves?

    A: No. Leave tracing to the police, who can seek subscriber and log details from the platform and service providers through legal process.

    This article is general information, not legal, financial or tax advice for any specific case. Lexfolio is a learning resource and does not provide legal services.

  • Cyber Crime Myths in India: 9 Mistakes Offices and Colleges Make

    Last updated: 12 November 2026

    Most offices and colleges learn cyber law the hard way: after the money has moved, the server is locked or a student’s photograph is circulating. By then the first hours have been spent on the wrong steps, because of beliefs that sound sensible but are wrong. This guide takes nine of those beliefs one by one and gives the correct position under Indian law as it stands on the date above.

    A quick note on the statutes: the main law is the Information Technology Act, 2000 (the IT Act). Ordinary offences such as cheating are now under the Bharatiya Nyaya Sanhita, 2023 (BNS), and police procedure is under the Bharatiya Nagarik Suraksha Sanhita, 2023 (BNSS), both in force since 1 July 2024.

    Learn this properly. Lexfolio offers Cyber Crime Law for Offices and Institutions, a self-paced certificate course on cyber crime law for offices and institutions. Course price: Rs 599.

    Myth 1: “We are too small or too ordinary to be targeted”

    Phishing e-mails, fake invoices and ransomware are sent in bulk, and a small office with weak passwords and one accounts clerk who can release payments is an easy target. The law also does not shrink with the size of the entity. The reporting directions issued by CERT-In (the Indian Computer Emergency Response Team) apply to service providers, intermediaries, data centres, body corporates and government organisations alike, so a private company or a college society that runs its own systems is covered.

    Myth 2: “A cyber incident is only a police matter, so we do not need to report it anywhere else”

    There are at least two separate reporting tracks, and the police track is the one people know. Under the directions issued by CERT-In on 28 April 2022 under section 70B(6) of the IT Act, listed incidents must be reported to CERT-In within six hours of noticing them. The list includes unauthorised access to IT systems or data, data breaches and leaks, ransomware and other malicious code, phishing and identity theft, attacks on e-mail, DNS or database servers, website defacement and unauthorised access to social media accounts. Reports go to incident@cert-in.org.in or the toll free number 1800-11-4949. CERT-In itself says that failing to comply can invite action under section 70B(7), which provides for imprisonment up to one year or a fine up to Rs 1 lakh, or both.

    The police track (a complaint or FIR) is separate. Doing one does not discharge the other.

    Myth 3: “The data law will tell us what to do after a breach, and it is not in force yet, so we can wait”

    Half of this is true and half is dangerous. The Digital Personal Data Protection Act, 2023 and the DPDP Rules, 2025 (published 13 November 2025) are being phased in. Only the first rules (definitions and the provisions on the Data Protection Board) took effect on 13 November 2025. The main duties of a data fiduciary, including security safeguards and breach intimation, are set to apply from 13 May 2027, eighteen months after publication. The Rules require a breach to be intimated to affected persons and to the Data Protection Board, with a detailed report to the Board within 72 hours.

    But “not yet in force” does not mean “no rule today”. Until then section 43A of the IT Act (compensation when a body corporate is negligent in protecting sensitive personal data) continues to apply, and the CERT-In six hour direction applies now. The DPDP Act will omit section 43A only when its section 44(2) comes into force on the same 13 May 2027 date. The DPDP Act’s schedule provides penalties of up to Rs 250 crore for failure to take reasonable security safeguards and up to Rs 200 crore for failure to give intimation of a breach.

    Myth 4: “Deleting the malicious file or wiping the machine solves it”

    It destroys the evidence. Cyber cases are won on logs, headers, device images and timestamps. The CERT-In directions already require service providers, body corporates and others covered to keep logs of all ICT systems for a rolling 180 days, stored within India, and to synchronise system clocks with NIC or NPL time servers or sources traceable to them. If you rebuild a server before the logs are copied, you may lose both the proof and the ability to show compliance. Isolate the machine from the network, do not switch off or wipe it, and copy logs first.

    Myth 5: “A screenshot is enough proof in court”

    A screenshot helps you complain, but electronic records are proved in a particular way. Under section 63 of the Bharatiya Sakshya Adhiniyam, 2023 (BSA), an electronic record produced as secondary evidence is admissible when accompanied by a certificate in the prescribed form, signed by the person in charge of the device or system and, in Part B, by an expert. The certificate carries a hash value of the record. In Pune Bar Association v. Union of India (22 May 2026) the Supreme Court upheld the section 63(4) certificate requirement as constitutional, describing a hash value as “synonymous with an electronic fingerprint”. The Court also read section 39 BSA as allowing a person with special skill in computer science and cyber forensics to qualify as an expert for the Part B certificate on the material before the court, and left the wider question open. So keep original files, exported logs and the device itself.

    Myth 6: “If we lose money to online fraud, the bank must refund us”

    It depends on how the money left the account. The Reserve Bank of India’s circular of 6 July 2017 on customer liability in unauthorised electronic banking transactions gives zero liability where the fault is the bank’s, and also where a third party breach occurs and the customer reports within three working days of the bank’s communication. If the customer reports after four to seven working days, liability is limited to the transaction value or a cap, whichever is lower. The caps are Rs 5,000 for BSBD accounts, Rs 10,000 for other savings accounts and for MSME and small individual current accounts (average balance or limit up to Rs 25 lakh), and Rs 25,000 for other current accounts. Beyond seven working days it depends on the bank’s board approved policy.

    The important limit: the circular deals with *unauthorised* transactions. If your own accounts officer, believing a fake invoice or a spoofed e-mail from a “vendor”, approved and released a payment, the transaction was authorised from the bank’s side and a refund right is generally much weaker. Report immediately anyway, because speed is what allows a bank to try to freeze the receiving account.

    Myth 7: “A cyber crime complaint is not an FIR, so nothing will happen”

    A complaint on the National Cyber Crime Reporting Portal (cybercrime.gov.in) or on the helpline 1930 is not by itself an FIR, but it can lead to one. Under section 173 of the BNSS, information about a cognizable offence can be given by electronic communication and is recorded when the informant signs it within three days. For large financial frauds, the Ministry of Home Affairs announced in May 2025 an e-Zero FIR system: complaints on the portal or 1930 with losses above Rs 10 lakh are converted automatically into a Zero FIR, which the complainant then follows up at the relevant cyber police station within three days. It was launched as a pilot in Delhi with plans for other states, so check what applies where you are. Below that threshold, do not wait for automatic conversion: file the portal complaint and also approach the local cyber cell or police station with a written complaint.

    Myth 8: “An employee who copies office data is only a disciplinary problem”

    It can be both a disciplinary and a criminal matter. Section 43 of the IT Act makes a person who, without permission of the owner, accesses, downloads, copies or extracts data from a computer liable to pay compensation. Section 66 turns the same acts into a crime where done dishonestly or fraudulently, with imprisonment up to three years or a fine up to Rs 5 lakh, or both. Section 72A punishes disclosure of personal information obtained under a lawful contract with intent to cause wrongful loss or gain, with imprisonment up to three years or a fine up to Rs 5 lakh, or both. Where an employee entrusted with data dishonestly misuses it, the offence of criminal breach of trust (section 316 BNS) may also be considered. Criminal liability depends on proof of intent, so preserve evidence and take advice before acting.

    Myth 9: “A fake account or photograph is a ‘social media problem’ that only the platform can fix”

    The platform is one route, not the only one. Identity theft using another person’s password or unique identification feature is punishable under section 66C (up to three years and a fine up to Rs 1 lakh). Cheating by personation using a computer resource is punishable under section 66D (up to three years and a fine up to Rs 1 lakh), and the personation limb of cheating also appears in section 319 BNS. Publishing a private image of a person without consent attracts section 66E (up to three years or a fine up to Rs 2 lakh, or both). Material showing children in sexually explicit acts falls under section 67B (up to five years on a first conviction). Use the platform’s grievance process, keep screenshots with URLs and dates, and file the portal complaint.

    What to do next: a one minute checklist

    • Name one person who decides what to do when something goes wrong, and keep the numbers where staff can find them.
    • Keep logs and back ups as the CERT-In directions expect, and know where they are stored.
    • Train staff on the two or three attacks that actually succeed: fake payment instructions, fake log in pages and password reuse.
    • Save the portal address (cybercrime.gov.in), the 1930 helpline and CERT-In’s reporting address in your incident plan.
    • Plan for the DPDP duties that begin on 13 May 2027.

    Learn this properly. Lexfolio offers Cyber Crime Law for Offices and Institutions, a self-paced certificate course on cyber crime law for offices and institutions. Course price: Rs 599.

    Frequently asked questions

    Do we have to report every spam e-mail to CERT-In?

    A: No. The six hour duty relates to the listed types of cyber incidents, such as unauthorised access, data breach, ransomware or compromise of systems. An ordinary spam message that was deleted unopened is not such an incident, but a successful phishing attack that gave an outsider access is.

    Does a college or school have to follow the CERT-In directions?

    A: The directions are addressed to service providers, intermediaries, data centres, body corporates and government organisations. Whether a particular institution falls in one of those categories depends on how it is constituted and what it operates, so confirm for your own entity. Government institutions are generally covered.

    How soon after a fraud should I call 1930?

    A: As soon as you notice it. The helpline exists so that banks and platforms can be alerted quickly to hold the money, and delay reduces the chance of recovery. Also tell your own bank in writing.

    Will the DPDP Act change what we have to do after a breach?

    A: Yes, from 13 May 2027 the data fiduciary duties, including breach intimation to the Data Protection Board and to affected persons, are scheduled to apply. Until then the IT Act and the CERT-In directions are what apply.

    This article is general information, not legal, financial or tax advice for any specific case. Lexfolio is a learning resource and does not provide legal services.

  • What to Do After a Cyber Attack or Data Breach at Your Office in India

    Last updated: 12 November 2026

    Short answer: if your office or institution has been hit by ransomware, a hacked mail account, a data leak or a payment fraud, treat the first day as a clock. Contain the problem without destroying evidence, report to CERT-In within six hours of noticing a listed incident, tell your bank at once if money moved, file a complaint with the police or the cyber crime portal, and keep a record of every step. The order below is a working checklist, not a substitute for advice on your own facts.

    This guide covers an organisation (company, firm, college, school, trust or government office) that controls its own systems. If you are an individual who lost money in a scam, the same ideas apply, but the six hour direction is aimed at organisations.

    Learn this properly. Lexfolio offers Cyber Crime Law for Offices and Institutions, a self-paced certificate course on cyber crime law for offices and institutions. Course price: Rs 599.

    The first day at a glance

    StepDo thisRule or source
    ———
    1Contain: isolate affected devices, change credentials, do not wipeEvidence preservation
    2Note the exact time you noticed the incidentCERT-In six hour clock runs from noticing
    3Report to CERT-In if the incident is of a listed typeDirections of 28 April 2022 under section 70B(6), IT Act
    4If money moved, call your bank and the helpline 1930RBI circular of 6 July 2017; National Cyber Crime Reporting Portal
    5File a complaint or FIRSection 173, BNSS
    6Preserve and certify electronic evidenceSection 63, BSA
    7Assess data protection dutiesIT Act section 43A now; DPDP Act and Rules from 13 May 2027
    8Fix, document, learnInternal record

    Step 1: Contain, but do not destroy

    Disconnect the affected computers from the network and the internet (pull the cable or switch off Wi-Fi rather than shutting the machine down, because shutting down can erase information held in memory). Change passwords and revoke access tokens from a clean device, starting with e-mail, banking and administrator accounts. Switch on multi-factor authentication where it was missing.

    Do not reinstall, reformat or delete anything yet. Make a list of devices, accounts and servers involved. If you use a vendor or cloud provider, ask them in writing to preserve logs. The CERT-In directions already expect covered entities to keep logs of all ICT systems for a rolling 180 days, stored within India, so the logs may exist even if the machine itself is rebuilt later.

    Step 2: Write down the time you noticed

    The six hour period under the CERT-In directions runs from the moment the entity notices the incident or is told about it. Record the time, who noticed it and how, in a simple log that you keep adding to. It protects you later if anyone asks when you knew.

    Step 3: Report to CERT-In if the incident is listed

    The directions issued by CERT-In on 28 April 2022 under section 70B(6) of the Information Technology Act, 2000 list incident types in an annexure. Those most relevant to offices and institutions include:

    • unauthorised access to IT systems or data, and compromise of critical systems or information
    • data breaches and data leaks
    • malicious code such as ransomware, spyware and trojans
    • identity theft, spoofing and phishing
    • attacks on mail, DNS or database servers and on network devices
    • website defacement or intrusion
    • unauthorised access to social media accounts
    • attacks affecting digital payment systems or cloud systems

    Report within six hours by e-mail to incident@cert-in.org.in or by phone on 1800-11-4949. The directions ask for the details of the incident as far as known, so send what you have and add more later. Failure to furnish information or comply can invite action under section 70B(7), which provides for imprisonment up to one year or a fine up to Rs 1 lakh, or both.

    Two related duties in the same directions are worth checking today and not on the day of an attack: keep system logs for 180 days within India, and synchronise your system clocks with the NIC or NPL time servers (or sources traceable to them).

    Step 4: If money has moved, act within minutes

    Call your bank’s fraud line and follow up in writing, giving the transaction reference, amount, time and the receiving account. Then report on the National Cyber Crime Reporting Portal (cybercrime.gov.in) or call the national helpline 1930. Speed matters because the receiving bank may be able to hold the funds only while they have not been moved on.

    Under the Reserve Bank of India’s circular of 6 July 2017, a customer has zero liability for an unauthorised electronic transaction where the fault lies with the bank, or where a third party breach occurs and the customer reports within three working days of the bank’s communication. After four to seven working days the customer’s liability is capped (Rs 5,000, Rs 10,000 or Rs 25,000 depending on the account type) and beyond seven working days it follows the bank’s board approved policy. The banks must credit the disputed amount provisionally within ten working days of notification. These protections concern *unauthorised* transactions. If your own staff released the payment on a fake instruction, expect a harder dispute and tell the bank exactly what happened.

    Step 5: File a complaint or FIR

    Take a written complaint to the cyber crime police station or the local police station, with copies of the evidence. Under section 173 of the Bharatiya Nagarik Suraksha Sanhita, 2023, information can also be given electronically; it is taken on record when the informant signs it within three days. For large financial frauds, the Ministry of Home Affairs announced in May 2025 an e-Zero FIR process under which complaints on the portal or 1930 involving more than Rs 10 lakh are converted automatically, starting as a pilot in Delhi. Whether it operates in your state may differ, so ask the police desk.

    What to include in the complaint:

    1. A short chronology with dates and times.
    2. Names and designations of the staff involved and what each did.
    3. Account numbers, transaction IDs, phone numbers, e-mail addresses, URLs and IP addresses seen.
    4. The loss, if any, and the date you informed the bank and CERT-In.
    5. A list of the evidence you are handing over or preserving.

    The offences usually considered are in the IT Act (for example sections 43 and 66 for unauthorised access and damage, 66C for identity theft and 66D for cheating by personation using a computer resource) and in the Bharatiya Nyaya Sanhita, 2023 (for example section 318 for cheating and section 319 for cheating by personation). The police decide which sections apply on the facts.

    Step 6: Preserve and certify the electronic evidence

    Copy, do not move, the originals. For each item note who copied it, when, from which device and with which tool. Keep e-mails in their original format with full headers, not only printed pages.

    In court, electronic records are proved under section 63 of the Bharatiya Sakshya Adhiniyam, 2023. A certificate in the prescribed form accompanies the record: Part A is given by the person in charge of the device or computer system and Part B by an expert, and the certificate records the hash value of the record. In Pune Bar Association v. Union of India, decided on 22 May 2026, the Supreme Court upheld the certificate requirement and described a hash value as “synonymous with an electronic fingerprint”. It also read section 39 of the BSA as allowing a person with special skill in computer science and cyber forensics to qualify as an expert for Part B on the material before the court, while leaving the wider question open. A forensic specialist engaged early can save you from a failed certificate later.

    Step 7: Check your data protection position

    Today, section 43A of the IT Act requires a body corporate that handles sensitive personal data to maintain reasonable security practices, failing which it can be liable to pay compensation to the person affected. The Digital Personal Data Protection Rules, 2025 were published on 13 November 2025 and the main obligations, including the duty to intimate a personal data breach, are scheduled to apply from 13 May 2027. Once they do, a data fiduciary will have to inform each affected person without delay, and the Data Protection Board with an initial description followed by a detailed report within 72 hours. The Act’s schedule sets maximum penalties of Rs 250 crore for failure to take reasonable security safeguards and Rs 200 crore for failure to give intimation of a breach.

    If the incident exposed personal data of students, patients, staff or customers, list what data, whose, and how many. You will need that list under any regime.

    Step 8: Close the loop

    • Reset all credentials, patch the entry point and test back ups before bringing systems back.
    • Brief staff in plain language on what happened and what to do next time.
    • Review contracts with IT vendors for incident support and log retention.
    • Keep a single folder with the incident log, reports to CERT-In, bank and police, and evidence certificates.
    • If you have cyber insurance, read the notice period in the policy and inform the insurer in time.

    What not to do: do not pay a ransom or “recovery agent” on the strength of a message alone, do not delete the attacker’s e-mails and do not contact the receiving account holder yourself. Leave contact with suspects to the police.

    Learn this properly. Lexfolio offers Cyber Crime Law for Offices and Institutions, a self-paced certificate course on cyber crime law for offices and institutions. Course price: Rs 599.

    Frequently asked questions

    What if I am not sure whether the incident is “listed” for CERT-In?

    A: The annexure covers most incidents an office is likely to face, including unauthorised access, phishing, ransomware and data leaks. If you are unsure, reporting is the safer course, and you can supplement the report as facts emerge.

    Is a complaint on cybercrime.gov.in enough, or do I also need an FIR?

    A: The portal complaint starts the process, but an FIR is what lets investigation and prosecution proceed formally. Follow up with the cyber crime police station and ask for the FIR number or a written acknowledgement.

    Can we investigate internally before telling anyone?

    A: You can begin internal checks, but the six hour period for listed incidents runs from noticing, and delay can cost recovery of money. Report first and investigate in parallel.

    Does an employee’s mistake, such as clicking a phishing link, make the employer liable?

    A: Liability depends on the facts: what safeguards the organisation had, what the contract with the affected person says and which law applies. Training and reasonable security practices are your best protection.

    This article is general information, not legal, financial or tax advice for any specific case. Lexfolio is a learning resource and does not provide legal services.

  • How to Write an AI Use Policy for Your Office in India: Step by Step

    Last updated: 12 November 2026

    Short answer: an AI use policy for an Indian office should do six things: say which tools are approved, sort information into classes and say which class may go into which tool, require a named human to check and own every output, set rules for personal data, say what to do when something goes wrong, and train staff and review the policy regularly. No Indian law prescribes a form for it, but the law you already have (contracts, the IT Act, copyright and, from 13 May 2027, the main duties under the data protection law) is exactly what the policy should translate into daily rules.

    This guide is for a small or medium employer, a department head or an HR or compliance person who has been asked to “write something on ChatGPT”. It is a drafting checklist with a skeleton you can adapt, not a ready made legal document.

    Learn this properly. Lexfolio offers AI at Work: Using ChatGPT and AI Tools Safely and Legally, a self-paced certificate course on using ChatGPT and other AI tools safely and legally at work. Course price: Rs 299.

    Why a written policy is worth the effort

    Staff are using these tools whether you permit them or not. A policy turns a vague worry into rules people can follow, and it gives the employer something to point to if an employee later breaches confidentiality or passes off unchecked output as fact. It also supports the position that the organisation took “reasonable security practices”, which is the standard section 43A of the IT Act uses today and the Digital Personal Data Protection Act, 2023 will use from 13 May 2027 when its main duties are scheduled to apply.

    The India AI Governance Guidelines issued by the Ministry of Electronics and Information Technology on 5 November 2025 favour voluntary measures and existing laws over a new AI statute, and encourage organisations that deploy AI to provide accessible grievance redressal. A written policy is the natural way to follow that approach.

    Step 1: Find out what is already being used

    Ask each team, without blame, which tools they use, on what accounts (company or personal), for what tasks and with what kinds of data. Check browser extensions, meeting note takers and tools built into software you already pay for. You cannot write rules for a practice you have not seen.

    Step 2: Classify your information

    A simple four class table is enough for most offices.

    ClassExamplesRule for public AI tools
    ———
    PublicPublished brochures, public rulesAllowed
    InternalDraft emails, internal notes with no personal dataAllowed on approved tools only
    ConfidentialClient contracts, pricing, unpublished results, source code, legal adviceNot allowed unless the tool is approved for it in writing
    Personal or sensitiveEmployee, student, patient or customer data; financial and health dataNot allowed; mask or anonymise first

    Add your own classes where a client or regulator already has one. If your clients’ contracts say that their data must not leave particular systems, those contracts override your internal table.

    Step 3: Approve tools and accounts

    Name the tools staff may use, on company accounts only. For each tool, someone should read the terms on four points: whether inputs are stored, whether they are used to improve the service, where data is processed and who owns the output. Ask for business or enterprise terms where confidential data is involved, and record the date you checked. Put a person in charge of approving new tools so that the question “can I use X?” has a clear answer.

    Step 4: Write the do and do not rules

    Keep them short enough to fit on one page.

    • Do use approved tools for drafting, summarising public material, brainstorming, translation drafts and formula help.
    • Do not enter confidential or personal information into an unapproved tool.
    • Do not present generated text, numbers or legal or medical statements as verified unless a person has checked them.
    • Do not generate or share images, voices or videos of real people without their written consent.
    • Do not use a tool to make or finalise decisions about hiring, pay, discipline or credit without a human reviewer.
    • Do not upload third party copyrighted material that you are not licensed to copy.

    Step 5: Make human review a rule, not a hope

    Name the role that is responsible for each type of output. The person who sends a document, signs an opinion or publishes a post owns it, whichever tool helped draft it. The Supreme Court’s recent remarks about non-existent citations in court filings (17 and 27 February 2026, and again on 5 May 2026) show the standard the system now expects: verification against authentic sources, and no excuse that “the tool made the mistake”. For important work, ask the reviewer to note what was checked and against which source.

    Step 6: Deal with personal data

    If you handle personal data, list which tools touch it and whether the vendor processes it on your behalf. From 13 May 2027 a data fiduciary will have to take reasonable security safeguards (the Rules mention measures such as encryption, masking or tokenisation, access controls, logging and monitoring, and back ups) and have contracts requiring processors to maintain equivalent safeguards. The Rules also provide for breach intimation to the Data Protection Board and to affected persons. Maximum penalties in the Act’s schedule are Rs 250 crore for failure to take reasonable security safeguards and Rs 200 crore for failure to give intimation of a breach. Build these into the policy now: approved tools only for personal data, a record of what is sent, and a clear no for anything not on the list.

    Step 7: Set up incident reporting

    Define what counts as an incident: confidential data pasted into an unapproved tool, an account taken over, a fake generated message used against the company. Give staff one person and one phone number to tell at once. Some events are also reportable outside the company. Under the CERT-In directions of 28 April 2022, listed incidents such as unauthorised access, data breaches and phishing must be reported to CERT-In within six hours of noticing. The directions also expect logs to be kept for 180 days within India.

    Step 8: Cover disclosure and intellectual property

    Decide when staff must say that a tool was used: for example in client work, submissions to authorities, academic or training material, and anything that carries your signature. For authorship, remember that section 2(d)(vi) of the Copyright Act, 1957 treats the person who causes a computer-generated work to be created as its author, but how this applies to text or images produced by a generative tool is not settled by an Indian court judgment, and the DPIIT working paper of 8 December 2025 on generative AI and copyright is only a consultation document. Do not promise clients exclusive ownership of generated material without checking the tool’s terms.

    Step 9: Train people and test the policy

    A one page policy nobody has read does nothing. Run a short session using real examples from your own office, run two or three test cases (for example, “can I paste this client agreement?”) and fix unclear rules. Give staff a place to ask questions without fear of blame.

    Step 10: Review at fixed intervals

    The law and the tools are both moving. The data protection obligations begin on 13 May 2027, the IT Rules were amended in February 2026 to require labelling of synthetic content, and courts and regulators continue to issue directions on AI. Put a review date in the policy, such as every six months, and name who will carry it out.

    A skeleton you can adapt

    1. Purpose and scope. This policy applies to all employees, contractors and interns using generative tools for work.

    2. Approved tools. Only tools on the approved list, used through company accounts.

    3. Information classes. Public, internal, confidential, personal. Confidential and personal information must not be entered into any tool unless approved in writing.

    4. Human responsibility. The person who uses or sends an output is responsible for checking it.

    5. Prohibited uses. Decisions about people without human review; synthetic images or voices of real people without consent; copying third party material without a licence.

    6. Incidents. Report to [name, phone] immediately.

    7. Disclosure. State tool use where a client, contract or rule requires it.

    8. Review. This policy is reviewed every six months.

    Have a lawyer review the final text against your contracts and sector rules before you adopt it.

    Learn this properly. Lexfolio offers AI at Work: Using ChatGPT and AI Tools Safely and Legally, a self-paced certificate course on using ChatGPT and other AI tools safely and legally at work. Course price: Rs 299.

    Frequently asked questions

    Is an AI use policy legally required in India?

    A: No Indian law requires a stand-alone one at present. But a policy helps show reasonable security practices and responsible governance, and certain contracts and regulators may expect one.

    Can we make employees sign the policy?

    A: Yes. Make acceptance part of onboarding and circulate updates with an acknowledgement. Whether a breach justifies disciplinary action depends on your service rules and the employee contract.

    Should we ban all AI tools instead?

    A: A complete ban is hard to enforce and pushes use onto personal accounts, where you have no visibility. A narrow approved list with clear data rules is usually safer.

    Do small offices need all ten steps?

    A: Not at full scale. A single page covering approved tools, data classes, human review and who to tell about an incident gives most of the protection.

    This article is general information, not legal, financial or tax advice for any specific case. Lexfolio is a learning resource and does not provide legal services.