💥Mains Ready By December. Smash Mains & Smash PYQ Admissions Open
Talk to Mentor

Civilsdaily

Login

GS Paper: GS3-17.Awareness in the fields of IT, Space, Computers, Robotics, Nano-technology, Bio-technology and issues relating to Intellectual Property Rights.

  • Saving faces: Use of facial recognition equipment at protest site is worrisome

    Why in the News

    The Delhi Police has told the Supreme Court that it deployed a facial recognition system at the site of the Cockroach Janta Party protests, along with a mobile surveillance van, a command and control vehicle, smart spectacles and drones. The disclosure came in the same proceeding. In that proceeding the force has continued to deny using excessive force or manhandling demonstrators, contrary to the protestors’ own testimonies. India is therefore normalising the technical ability to subject political gatherings to searchable biometric surveillance. Legislation and judicial oversight have not yet settled when the state may lawfully do so. The contest is between a policing capability that is already operational and a legal framework that names no threshold, no authorising authority and no retention rule for its use.

    What is a facial recognition system?

    1. It converts a face into a searchable record: The system extracts measurable geometric features from a face image and stores them as a numeric template that can be matched against other templates.
    2. Two distinct operations, two different risk profiles: Verification matches one face against one claimed identity. Identification matches one face against an entire database, and only the second turns a crowd into a search.
    3. Real time capture removes the choice to participate anonymously: Cameras enabled with Artificial Intelligence (AI) scan faces as people move and run matches against a database without any interaction with the person scanned.

    What surveillance equipment did the police say it had deployed?

    1. A facial recognition system with AI enabled cameras: These scanned faces in real time against a database. The demonstration was under way at the time.
    2. A mobile surveillance van: A vehicle mounted capture platform able to move with the crowd rather than covering a fixed field.
    3. A command and control vehicle: The on site node where feeds from the various capture devices were aggregated and acted on.
    4. Smart spectacles: Wearable devices used to identify individuals on the move, which extends identification beyond fixed and vehicle mounted cameras.
    5. Drones and videographers: Aerial and handheld recording covering the site from angles the ground cameras did not reach.
    6. Private contractors hold two of these systems: The van and the spectacles have been tied to private contractors on terms that have never been disclosed, so a commercial entity sits inside the capture chain on an unknown mandate.

    What did the police not disclose?

    1. Whether every face in range was processed: The force has not addressed whether actual biometric processing occurred for every individual within the range of the cameras, which is the difference between targeted identification and mass capture.
    2. Where discarded images went: Images from checks that produced no match were open to copying in the interim, and the force has not said whether any copy survives.

    Which laws currently govern facial recognition, and what do they leave open?

    1. No statute governs the technology: There is no law in force that regulates the use of facial recognition systems by the state, so deployment rests on executive decision alone.
    2. The data protection law is not yet operative on this point: The Digital Personal Data Protection Act, 2023, whose data processing obligations are not yet in force, still makes broad exemptions for state agencies.
    3. The existing police database is purpose limited: The Automated Facial Recognition System of the National Crime Records Bureau is meant for identifying criminals and unidentified bodies, not for scanning an assembly.
    4. The 2022 statute widened records, not subjects: The Criminal Procedure (Identification) Act, 2022 expanded the set of records the police may collect, but only from specified persons rather than from the public at large.
    5. The gap is the crowd: Every one of these instruments operates against identified groups of people, and none of them authorises indiscriminate capture of everyone present at a location.

    Can mass biometric capture at a protest survive the proportionality test?

    1. The state carries the burden: Interference with the right to privacy must clear a well established proportionality test, and the burden of establishing each limb sits on the state.
    2. The first limb already fails on the facts: The existence of a legitimate objective is hard to establish for facial recognition used en masse, because the technology is indiscriminate at the point of capture and cannot be aimed at a suspect.
    3. A less restrictive alternative exists: Conventional policing achieves the same objective of maintaining order and identifying offenders without capturing the biometrics of every person present.
    4. Constitutional validity is not the only test: Even leaving aside the constitutionality of the police action, a capability deployed without a governing standard sets the precedent for the next deployment.

    How does biometric surveillance affect the right to protest?

    1. The deterrent operates before any legal restriction: The chilling effect on potential participants curtails the right to protest without any order prohibiting the protest.
    2. Anonymity is part of the freedom: Assembly has historically carried the assurance that presence in a crowd is not the same as being recorded as an individual participant.
    3. The cost falls on people with the most to lose: Government employees, students facing institutional discipline and people in precarious work self select out once presence becomes a permanent identified record.
    4. The chill is unfalsifiable: Nobody counts the people who stayed home, so the harm never appears in the record a court would examine.

    What remains unanswered before this use can be justified?

    1. The access controls: It is unknown which officers, agencies or contractor personnel could query the captured images and against which databases.
    2. The authorising legal provisions: The provisions relied on to authorise the major decisions, including the decision to scan an entire assembly, have not been identified.
    3. The false positive rate: The expected error rate of the system has not been stated, and a false match at a protest site produces detention of an innocent person on machine evidence.

    Challenges to the regulation of facial recognition technology in India

    1. Deployment has run far ahead of legislation: State and city police forces have procured systems under general policing powers rather than under any enabling statute. Eg. Several State police departments and airports adopted facial recognition without a dedicated legal framework in place. Fix. Enact a facial recognition statute prescribing permitted purposes, a judicial or independent authorisation requirement, and a fixed retention period.
    2. The data protection statute exempts the principal user: Broad exemptions for state agencies mean the very actor conducting mass capture falls outside the consent and purpose limitation architecture. Eg. The Justice B N Srikrishna Committee had recommended narrow and specified exemptions rather than open ended ones on grounds of sovereignty and public order. Fix. Replace the blanket agency exemption with a case by case exemption that must be notified with reasons and laid before Parliament.
    3. Accuracy is unequal across populations: Error rates for facial recognition are consistently higher for darker skinned faces, women and younger subjects, so the burden of false matches is not evenly distributed. Eg. Independent testing of commercial systems has repeatedly found the highest error rates for darker skinned women. Fix. Mandate published accuracy testing disaggregated by skin tone, sex and age before any system is procured for policing use.
    4. Private contractors sit inside the state’s capture chain: Outsourcing capture hardware and processing places biometric data with entities that are not accountable through public law remedies. Eg. Police facial recognition deployments in several States run on vendor supplied platforms whose procurement contracts are not in the public domain. Fix. Require every surveillance procurement contract to be published with its data handling clauses, and make the contractor a joint respondent in any privacy proceeding.
    5. There is no oversight body with jurisdiction: No standing authority audits police biometric systems, so no institution can verify retention, deletion or match logs after the event. Eg. Agencies conducting interception under existing law are reviewed only by an internal executive review committee. Fix. Establish a statutory surveillance oversight commission with power to inspect match logs and order deletion.
    6. Function creep is the default trajectory: A database built for one purpose is progressively opened to others once the infrastructure exists. Eg. Facial recognition adopted for airport boarding convenience has been proposed for wider identity verification uses. Fix. Write a statutory bar on cross purpose querying, with each authorised purpose requiring a separate legislative amendment.

    Conclusion

    The disclosure establishes that the capability to convert a political gathering into a searchable biometric record is already deployed, contracted out in part, and operating without a statute that says when it may be used. The proportionality test, on the facts available, is not close: the technology captures indiscriminately, a less restrictive alternative exists, and the state has not identified the provision that authorised the decision. Until Parliament enacts a facial recognition law with a stated purpose, an authorising authority, a retention limit and published accuracy standards, each deployment simply widens the precedent for the next one.

    “[2024, GS3, 10 marks] Describe the context and salient features of the Digital Personal Data Protection Act, 2023″

  • How will Gaganyaan’s thermal shield protect the crew?

    Why in the News

    The Gaganyaan crew module will hit the atmosphere at 7,500 to 8,000 metres per second on return, with its exterior reaching 1,800 degrees Celsius while the structure must stay below 150 degrees Celsius. The shield chosen to hold that gap is a sacrificial ablative layer 30 to 35 millimetres thick, a choice driven by the mission’s single use design and India’s own re entry heritage rather than by peak performance.

    What is a thermal protection system?

    1. What it does: A thermal protection system is the outer layer that keeps a re entering vehicle’s structure and interior within survivable temperature while its exterior is exposed to the heat of atmospheric entry.
    2. Why it is needed: Almost all of the crew module’s kinetic energy is dissipated into the atmosphere as heat energy, and the small portion directed back towards the module is still intense enough to melt it.
    3. What it protects: It maintains the module’s structural integrity and keeps the interior within the temperature limit the structure and the crew can tolerate.
    4. How it is classified: Systems are grouped by how they remove heat, into ablative, radiative and heat sink types.

    What is heat flux?

    1. Definition: Heat flux is the rate at which heat energy passes through a unit area of a surface, measured in watts per square metre.
    2. Why it varies on a capsule: It is highest at the point of the vehicle that meets the airflow first, which is why the nose cap carries the most demanding shield material.

    What is a boundary layer?

    1. Definition: The boundary layer is the thin region of gas immediately next to a moving vehicle’s surface, where the flow is slowed by contact with that surface.
    2. Why it matters in ablation: Gases escaping from the decomposing shield thicken and cool this layer, which blocks intense heat from being transferred into the module.

    Why is atmospheric re entry harder than ascent for a crewed mission?

    1. Ascent is controlled and gradual: A rocket accelerates slowly through the atmosphere on the way up specifically to keep the mechanical loads on the vehicle to a minimum.
    2. Re entry cannot be aborted: Once the descent begins there is no provision to abort the mission, so every system must work through to splashdown.
    3. The crew cannot intervene: There is only a limited role for the crew to intervene and correct any system non conformance during descent.
    4. The event is too fast for human correction: Atmospheric descent is incredibly fast and the deceleration forces change constantly, and human response times are simply too high to manually correct a sudden system abnormality.
    5. What follows from this: All systems must therefore be made robust enough to withstand the scorching conditions of re entry on their own, since design margin substitutes for intervention.

    What thermal conditions must the Gaganyaan crew module survive?

    1. Entry velocity: The crew module will hit the atmosphere at a speed of 7,500 to 8,000 metres per second on return from its orbit around the earth.
    2. Energy dissipation: More than 99 per cent of that kinetic energy will be dissipated into the atmosphere as heat energy.
    3. Exterior temperature: The exterior of the module will encounter temperatures as high as 1,800 degrees Celsius in some regions.
    4. Shield thickness: The thermal protection system is just 30 to 35 millimetres thick.
    5. Interior limit: That layer must keep the module’s temperature safely below 150 degrees Celsius while performing the task of maintaining structural integrity.

    How do ablative, radiative and heat sink systems each remove heat?

    1. Ablative: A single use system that removes heat energy by sacrificing its own layers through chemical and physical processes, absorbing extreme quantities of thermal energy and chemically decomposing into a protective layer of solid char and outgassing vapours.
    2. The decomposition physically carries heat away from the module as the material burns off, and the escaping gases create a cooler boundary layer that blocks heat transfer into the module.
    3. Carbon phenolic and silica phenolic are examples of ablative materials.
    4. Radiative: A system that absorbs the extreme heat of re entry and then releases it back into space as electromagnetic radiation, primarily in the infrared spectrum and also as visible light when it is extremely hot.
    5. It remains intact and withstands the heat without melting or degrading, which makes it suited to reusable re entry vehicles.
    6. Heat sink: A system that absorbs heat energy and raises its own temperature without melting or changing phase in any other way.
    7. Copper and aluminium are examples of heat sink materials.

    Why has the Indian Space Research Organisation chosen an ablative shield for the crew module?

    1. It matches the mission’s design philosophy: The Gaganyaan crew module is a single use vehicle, and an ablative system is a single use system, so the shield’s life and the module’s life are the same.
    2. It is proven and robust: The Indian Space Research Organisation (ISRO) has selected it as a proven and highly robust solution rather than the highest performing one available.
    3. It tolerates fluctuating heat loads: Ablative heat shields can easily handle fluctuating heat loads to protect the structure underneath, which matters when the descent profile varies.
    4. Radiative systems are less forgiving: Any design error in a radiative system can quickly cause dangerous overheating, so its margin for error is narrower.
    5. It avoids a maintenance burden: An ablative system withstands an extreme thermal load without requiring complex or delicate surface maintenance between flights.
    6. It avoids the reusable system’s cost structure: By avoiding the expensive manufacturing, specialised inspection and complex installation processes associated with a reusable radiative system, ISRO has taken the safer and more cost effective option.

    Does choosing a single use shield trade away reusability for safety?

    1. What is given up: A sacrificial shield is consumed on every flight, so a new heat shield must be manufactured and installed for each mission rather than inspected and reflown.
    2. The recurring cost consequence: Per flight cost stays flat across a programme instead of falling with flight rate, which is the opposite of the economics a high cadence programme needs.
    3. Why the trade is correct for this mission: Reusability only pays back over a high flight rate, and a first generation crewed programme flying occasional missions never reaches that rate.
    4. Where the trade stops working: A sustained crew rotation programme to an orbital station changes the flight rate, at which point the reusable radiative option becomes the economically relevant one.
    5. The safety side of the trade: The ablative system’s tolerance of fluctuating heat loads and its independence from surface inspection are precisely the properties a programme flying its first crew needs most.

    What does India’s own re entry heritage contribute to the Gaganyaan shield?

    1. The first re entry mission: The Space Capsule Recovery Experiment, India’s maiden re entry mission, used a carbon phenolic ablative to protect the module’s nose cap, where heat flux was the highest.
    2. The crew module demonstration: The Launch Vehicle Mark-3 (LVM3) flew the Crew Module Atmospheric Re-entry Experiment (CARE) in 2014. That flight successfully demonstrated crew module re entry using an ablative thermal protection system.
    3. What that established: The 2014 mission established the foundational technology that is now being used in the Gaganyaan programme, so the shield is an inheritance rather than a new development.
    4. Why heritage reduces risk: Material characterisation, manufacturing process and flight data already exist for the ablative route, which removes the qualification uncertainty a new material class would carry.
    5. The programme position: The Gaganyaan crew module is built on this ablative heritage and on the lessons learned from both earlier missions.

    What does the SpaceX Crew Dragon comparison show about ablative shield design choices?

    1. United States, the Crew Dragon shield: The Crew Dragon capsule of SpaceX uses an ablative material named phenolic impregnated carbon ablator, or PICA, a lightweight carbon fibre matrix filled with a phenolic resin.
    2. The shared design logic: A crewed capsule operator with a very different cost structure has arrived at the same ablative class of solution, which indicates the choice follows from the capsule form rather than from budget constraint.
    3. The design feature that differs: PICA’s lightweight carbon fibre matrix trades density for mass saving, while carbon phenolic of the kind flown on India’s first re entry mission is denser and carries higher heat flux at the nose.
    4. The limit of this comparison: This is the single foreign system named in the evidence here, so it establishes that ablative shielding is the standard choice for crewed capsules, not a ranked comparison of national capsule programmes.

    Challenges to the Gaganyaan thermal protection system

    1. Ground testing cannot reproduce full re entry: No ground facility reproduces the combined velocity, heat flux and duration of an orbital re entry, so qualification relies on partial simulation and analysis. Eg. Arc jet plasma facilities test coupons at representative heat flux but not at the full 7,500 to 8,000 metres per second entry velocity.
    2. Bond line integrity over a curved surface: A 30 to 35 millimetre layer must adhere uniformly over the module’s full curvature, and a bond defect creates a local hot path into the structure. Eg. Shuttle era thermal protection failures originated in localised damage to the protective layer rather than in the material’s bulk performance.
    3. Predicting the recession rate: Ablative design depends on predicting how much material burns off, and an over prediction adds dead mass while an under prediction risks burn through. Eg. Nose cap regions carry the highest heat flux and therefore the largest uncertainty in recession estimates.
    4. Mass penalty on the launch vehicle: A sacrificial shield sized with margin is heavy, and every kilogram of shield reduces the payload the human rated launcher can carry. Eg. The human rated LVM3 has to lift the crew module, service module and shield together to a 400 kilometre orbit.
    5. Manufacturing repeatability: Each mission needs a newly manufactured shield, so process variation between production batches becomes a flight safety variable rather than a quality issue. Eg. Carbon phenolic layup is a manual intensive process where resin content and fibre orientation must be reproduced identically each time.
    6. Recovery environment after splashdown: A charred shield must survive water impact and sea recovery without compromising the crew compartment. Eg. India’s first re entry mission was recovered from the Bay of Bengal, which is the recovery zone the crewed programme also plans to use.
    7. Single point criticality: With no abort provision once descent begins and limited crew intervention, the shield has no backup system to fall back on. Eg. Human response times are too high to correct a sudden thermal abnormality during a descent where deceleration forces change constantly.

    Conclusion

    The Gaganyaan crew module’s protection against a 1,800 degrees Celsius re entry rests on a 30 to 35 millimetre ablative layer that sacrifices itself to carry heat away and hold the structure below 150 degrees Celsius. The choice of an ablative over a radiative system follows from the module’s single use design, its tolerance of fluctuating heat loads and the technology base established by India’s first re entry mission and the 2014 crew module demonstration. The programme’s current status is that the shield is qualified on this heritage, with the first uncrewed test flight launching shortly.

    Human Spaceflight Programme of India

    1. What it is: Gaganyaan is India’s human spaceflight programme, aimed at demonstrating the capability to launch a crew to low earth orbit and return them safely to Indian waters.
    2. Mission profile: The mission is designed to carry a crew of up to three to an orbit of about 400 kilometres for a mission duration of up to three days, followed by splashdown recovery.
    3. The launch vehicle: The launcher is a human rated version of the LVM3, designated the Human rated Launch Vehicle Mark-3 (HLVM3), modified with additional redundancy and a crew escape system.
    4. The orbital module: The crew module and the service module together form the orbital module, with the crew module being the pressurised habitable segment that returns.
    5. Institutional base: The Human Space Flight Centre was established at Bengaluru in 2019 to lead the programme, with the Vikram Sarabhai Space Centre responsible for launch vehicle and re entry systems.
    6. The longer roadmap: India’s stated goals extend to the Bharatiya Antariksh Station by 2035 and a crewed lunar landing by 2040.

    Laws and Treaties Governing Space Activities

    1. Outer Space Treaty, 1967: Makes States internationally responsible for national space activities, whether carried on by governmental or non governmental entities, and bars national appropriation of outer space.
    2. Rescue Agreement, 1968: Obliges States to assist astronauts in distress and to return them and any recovered space objects to the launching authority.
    3. Liability Convention, 1972: Makes a launching State absolutely liable for damage caused by its space object on the surface of the earth or to aircraft in flight.
    4. Registration Convention, 1975: Requires launching States to maintain a national registry of space objects and to furnish details to the United Nations.
    5. Moon Agreement, 1979: Declares the Moon and its resources the common heritage of mankind, and India has signed but not ratified it.
    6. Indian Space Policy, 2023: Defines the roles of ISRO, the Indian National Space Promotion and Authorisation Centre, NewSpace India Limited and non governmental entities in the Indian space ecosystem.
    7. Space Activities Bill, 2017: A draft domestic law to license and regulate private space activity in India, which was circulated for comment and never enacted.
    8. Satellite Communications Policy and spectrum rules: Govern authorisation of satellite services, with spectrum assignment handled under the Telecommunications Act, 2023.

    “[2025] Consider the following space missions:

    I. Axiom-4

    II. SpaDeX

    III. Gaganyaan

    How many of the space missions given above encourage and support microgravity research?

    (a) Only one

    (b) Only two

    (c) All the three

    (d) None

  • China lands a rocket first stage for the first time with Zhuque-3

    Why in the News

    China has recovered the first stage of a rocket on land for the first time, using the reusable rocket Zhuque-3, which was launched on Wednesday morning. It is the country’s second rocket stage recovery overall, after a sea platform recovery in July, and the first to use deployable landing legs. State media described the result as a major breakthrough in the country’s reusable rocket technology.

    What is a reusable rocket?

    1. About: A reusable rocket is a launch vehicle whose stages are recovered intact after flight and flown again, instead of being discarded once the payload is delivered.
    2. Why it lowers cost: The first stage carries most of the engines and structure, so recovering it avoids rebuilding the most expensive part of the vehicle for every launch.
    3. How recovery works: The stage separates after boost, reorients, uses engine burns to slow its descent and lands vertically on a pad or on a sea platform.
    4. What landing legs add: Deployable landing legs stabilise the stage at touchdown on ground, which is why their first use is treated as a distinct technical milestone.

    What did the Zhuque-3 flight achieve?

    1. Launch and recovery: Zhuque-3 was launched on Wednesday morning and its first stage was recovered afterward.
    2. First on land: This marks China’s first successful recovery of a rocket first stage on land.
    3. Second overall: It is the second time the country has recovered a rocket stage, following a successful recovery on a sea platform in July.
    4. New hardware: The recovery marked China’s first use of deployable landing legs.
    5. Official assessment: The state news agency deemed the result a major breakthrough in the country’s reusable rocket technology.

    How does this compare with earlier recoveries?

    1. China’s July recovery: On 10 July, the first stage of a Long March-10B rocket separated from the second stage after lift off and returned to a platform at sea.
    2. The difference land makes: A sea platform recovery avoids overflight of populated areas, while a land recovery removes the need for a recovery vessel and shortens the turnaround.
    3. United States, SpaceX: SpaceX has been recovering rockets since 2015 and has driven down launch costs by reusing hardware that would otherwise be discarded after carrying satellites and other payloads toward space.
    4. United States, Blue Origin: Blue Origin has likewise been recovering boosters since 2015, establishing vertical landing as a repeatable rather than experimental technique.
    5. What the comparison shows: China is closing a capability gap that has stood for a decade, and the operator here is a private launch company rather than the state programme.

    Why does reusability decide launch economics?

    1. Cost per launch: Reuse spreads the cost of building a stage across several flights, which is the single largest lever on the price of access to orbit.
    2. Launch cadence: Recovery shortens the interval between flights, which matters for deploying large satellite constellations.
    3. The payload penalty: Propellant reserved for the landing burn and the mass of legs and grid fins reduce the payload the same vehicle can carry.
    4. The break even condition: Reuse pays only when the same stage flies many times, so refurbishment cost and inspection time determine whether the saving is real.
    5. Strategic consequence: Cheaper and more frequent launch capacity translates directly into faster deployment of communication, navigation and remote sensing assets.

    Conclusion

    Zhuque-3’s flight gives China its first land recovery of a rocket first stage and its second stage recovery in six weeks, after the Long March-10B sea platform recovery of 10 July. The flight also carried the country’s first use of deployable landing legs, which is the hardware element that makes routine ground landings possible. The state news agency has called it a major breakthrough in reusable rocket technology. The next measure of the achievement is whether the recovered stage is refurbished and reflown, since recovery without reflight does not deliver the cost saving that reusability exists to produce.

    “[2016] What is ‘Greased Lightning-10 (GL-10)’, recently in the news?

    (a) Electric plane tested by NASA

    (b) Solar-powered two-seater aircraft designed by Japan

    (c) Space observatory launched by China

    (d) Reusable rocket designed by ISRO

  • Teen ChatGPT: Safety Moves to Age Verification

    Why in the News

    OpenAI is rolling out a separate version of ChatGPT for teenagers, with tighter restrictions on conversations about self harm, suicide, eating disorders and sexual content. The move follows cases in which teenagers who died by suicide had interacted extensively with chatbots beforehand, and it arrives while a United States Federal Trade Commission (FTC) inquiry into seven AI companies is under way. The safeguard depends on estimating a user’s age and on parents choosing to switch controls on, which are the two weakest links in the chain.

    What is ChatGPT for Teens?

    1. About: It is a more restricted version of ChatGPT into which OpenAI places users it identifies as being under 18.
    2. Content limits: The teen version avoids romantic or sexual conversations and places stronger limits around self harm related content.
    3. Anti anthropomorphism rule: The chatbot is discouraged from presenting itself as conscious or as emotionally attached to the user.
    4. Parental layer: Parents can link their accounts, set usage restrictions and receive alerts in certain situations.

    How does OpenAI decide who is a teenager?

    1. Signal based estimation: OpenAI uses a combination of signals to estimate whether a user could be under 18.
    2. The three signals named: How an account is used, the subjects discussed in it, and how long the account has existed.
    3. Override of stated age: ChatGPT can automatically place an account under the teen safeguards even if a different age was entered at signing up.
    4. The admitted limit: Age detection systems are not foolproof, and several parental controls depend on families opting in.
    5. What that makes the product: The teen version is an attempt to reduce some of the risks emerging around AI companionship rather than a complete fix for them.

    Why do AI chatbots pose a different risk from social media?

    1. Designed agreeability: AI chatbots are programmed to be agreeable companions that validate users’ feelings, which is not how conventional social media platforms operate.
    2. The harm pathway: Unchecked validation can intensify suicidal behaviour and self mutilation among vulnerable children confiding their deepest fears.
    3. Neurological vulnerability: Children’s developing brains make them particularly vulnerable to AI systems that create dopamine responses.
    4. The combination that matters: The technology is highly responsive, anthropomorphic and adept at mimicking empathy, and for adolescents still developing judgement and a sense of self, that combination proves pernicious.
    5. Everyday embedding: For a generation of digital natives, AI is already a sounding board for curiosity and a companion shaping how they learn, communicate and seek reassurance.

    What do the litigation and the studies show?

    1. The Adam Raine suit: The parents of 16 year old Adam Raine sued OpenAI last year, alleging that ChatGPT had validated his suicidal thoughts and discussed methods of self harm before his death in April 2025.
    2. OpenAI’s own admission: The company acknowledged that some of its safety protections could become less reliable over the course of long conversations.
    3. The Character.AI settlement: Character.AI and Google agreed this year to settle a lawsuit filed by the mother of a 14 year old who died by suicide in 2024 after extensively interacting with a Character.AI chatbot, with the mother alleging he had developed an intense emotional attachment to the bot.
    4. The 2025 United States study: It found that ChatGPT provided dangerous responses to teens discussing self harm, substance abuse and eating disorders, including drafting suicide letters.
    5. A second study: It found chatbots suggesting violence, self harm and substance use every five minutes during testing.

    What are regulators elsewhere doing?

    1. United States, Federal Trade Commission: The FTC, the country’s consumer protection and competition regulator, has opened an investigation into seven AI companies, including OpenAI, over the effects of their products on children.
    2. United States, the Meta trial: Meta is facing a trial on the ground that it deliberately designed Facebook and Instagram to exploit young users’ vulnerabilities and to make its platforms addictive.
    3. China: It has moved to restrict AI systems that encourage emotional dependence, targeting the companionship design itself rather than the content output.
    4. India: It relies on a patchwork of laws, regulations and platform led interventions rather than a dedicated instrument for AI and minors.
    5. What the set demonstrates: Two jurisdictions are acting through litigation and inquiry after the harm, one is acting on product design in advance, and India has neither route settled.

    Why is a safer chatbot not the same as a safe one?

    1. Age prediction is an estimate: The safeguard applies only once the system correctly guesses that the user is a minor, and children can misrepresent their age.
    2. Enforcement dependent restrictions: Content restrictions are only as effective as their enforcement, which is not independently observable from outside the company.
    3. Track record on earlier controls: OpenAI introduced parental controls last year, and critics quickly demonstrated that these could be easily bypassed.
    4. Reactive sequencing: The protections arrived only after sustained public and legal pressure, which is a reminder that children’s online safety cannot be left to Big Tech alone.
    5. The tension that remains: A company that profits from engagement is being asked to design against the very property, unconditional validation, that generates the engagement.

    Challenges to AI Safeguards for Minors

    1. Unverifiable age estimation: The safeguard triggers on inference rather than on verified identity. e.g. OpenAI relying on account usage patterns, discussed subjects and account age to guess whether a user is under 18.
    2. Opt in dependence: Protections that require a parent to activate them reach only supervised households. e.g. the parental controls introduced last year that critics demonstrated could be easily bypassed.
    3. Safety degradation over long sessions: Guardrails hold in short exchanges and weaken in the extended conversations minors actually have. e.g. OpenAI’s acknowledgement that some protections become less reliable over the course of long conversations.
    4. Cross platform substitution: A restriction on one service pushes the user to a less restricted one. e.g. Character.AI, whose chatbot featured in the 2024 death that Google and the company settled this year.
    5. Absence of independent testing: Only external researchers have surfaced the failure modes, and they have no standing access. e.g. the 2025 United States study that found ChatGPT drafting suicide letters for teens.
    6. Divergent national rules: A globally distributed product faces incompatible obligations across markets. e.g. China restricting emotionally dependent AI systems while India relies on a patchwork of laws and platform led interventions.
    7. No liability standard for conversational harm: Existing intermediary law was written for hosted content, not for generated responses. e.g. the Adam Raine suit, which turns on whether a chatbot’s own outputs contributed to a death.

    Conclusion

    The property that makes chatbots compelling for adolescents, unconditional and empathetic sounding validation, is the same property that turned them dangerous in the Raine and Character.AI cases. ChatGPT for Teens restricts content, discourages the bot from claiming emotional attachment and adds parental linkage, which is a welcome and overdue intervention. It nonetheless rests on age estimation that is admittedly not foolproof and on controls that families must opt into, after the previous generation of parental controls was shown to be bypassable. What remains missing is independent testing, transparency and external scrutiny, alongside digital literacy for the parents the safeguards assume will be watching.

    Child Online Safety and Artificial Intelligence Governance in India

    1. About: Child online safety covers the protection of minors from harmful content, exploitative design, data exploitation and psychological harm arising from digital products.
    2. The distinctive AI risk: Generative systems produce responses rather than host content, so harm arises from the model’s own output and not from a third party post an intermediary can be asked to take down.
    3. Companionship design: Systems built to maximise engagement through empathy simulation create attachment, which is why regulation is beginning to target design features rather than only content categories.
    4. India’s scale: India has one of the world’s largest populations of internet users under 18, with smartphone access typically arriving before any formal digital literacy instruction.
    5. Regulatory posture: India has no dedicated artificial intelligence statute, and obligations flow from the Information Technology Act, 2000, data protection law and platform self regulation.
    6. Institutional anchor: The National Commission for Protection of Child Rights is the statutory body that issues advisories and takes cognisance of child rights violations, including online ones.

    Laws and Rules Governing Children’s Online Safety in India

    1. Information Technology Act, 2000: The parent statute for electronic records, intermediary liability and cyber offences.
    2. Section 79: Grants intermediaries conditional safe harbour subject to due diligence, which is the hook for content obligations.
    3. Section 67B: Penalises the publication and transmission of material depicting children in sexually explicit acts.
    4. Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021: Impose due diligence, grievance redress and expeditious removal obligations on intermediaries and significant social media intermediaries.
    5. Digital Personal Data Protection Act, 2023: Defines a child as a person below 18 and requires verifiable parental consent before processing a child’s personal data.
    6. Design prohibition: Bars tracking, behavioural monitoring and targeted advertising directed at children.
    7. Protection of Children from Sexual Offences Act, 2012: Criminalises sexual offences against children, including the use of children for pornographic purposes.
    8. Juvenile Justice (Care and Protection of Children) Act, 2015: Provides the care, protection and rehabilitation framework for children in need of care.
    9. Commissions for Protection of Child Rights Act, 2005: Establishes the National and State Commissions for Protection of Child Rights with powers of inquiry into violations.
    10. Bharatiya Nyaya Sanhita, 2023: Carries the general criminal provisions on abetment of suicide and obscenity that apply where a digital product is alleged to have contributed to harm.

    Government Initiatives for Child Online Safety

    1. IndiaAI Mission: The national programme for compute, datasets, applications and a safe and trusted artificial intelligence pillar covering risk assessment and governance tools.
    2. Cyber Crime Prevention against Women and Children scheme: Funds State capacity for handling online offences against women and children, including forensic and training support.
    3. National Cyber Crime Reporting Portal: Provides a dedicated reporting channel for child sexual abuse material and other online offences.
    4. Information Security Education and Awareness programme: Runs cyber safety awareness for students, teachers and parents through the Ministry of Electronics and Information Technology.
    5. Indian Computer Emergency Response Team advisories: Issues public advisories on online safety practices and coordinates incident response.
    6. National Commission for Protection of Child Rights advisories: Issues directions to platforms on age assurance, harmful content and child data practices.
    7. Cyber Swachhta Kendra: Operates as the botnet cleaning and malware analysis centre supporting safer end user devices.

    Key Facts about Children and the Digital Environment

    1. The Digital Personal Data Protection Act, 2023 sets the threshold for a child at below 18 years, which is higher than the 13 year threshold under the United States Children’s Online Privacy Protection Act, 1998.
    2. The National Commission for Protection of Child Rights is a statutory body constituted under the Commissions for Protection of Child Rights Act, 2005.
    3. Safer Internet Day is observed on the second Tuesday of February.
    4. The European Union Artificial Intelligence Act, 2024 is the first comprehensive statute to classify artificial intelligence systems by risk tier and to ban specified manipulative practices.
    5. The Convention on the Rights of the Child, 1989, to which India is a party, requires protection of children from all forms of exploitation prejudicial to their welfare.
    6. General Comment No. 25 (2021) of the United Nations Committee on the Rights of the Child extends child rights obligations explicitly to the digital environment.

    Challenges in Regulating Artificial Intelligence Use by Children

    1. Verifiable parental consent at scale: The law demands verification without prescribing a workable method that does not itself collect more child data. e.g. the Digital Personal Data Protection Act, 2023 requiring verifiable parental consent for every under 18 user.
    2. Mismatched age thresholds: A global product faces a different definition of a child in each market. e.g. India setting the threshold at 18 while the United States Children’s Online Privacy Protection Act, 1998 sets it at 13.
    3. Attributing harm to a model output: Causation is contested when the alleged harm is a conversation. e.g. the Adam Raine suit and the Character.AI settlement, both of which turn on whether chatbot responses contributed to a death.
    4. Cross border enforcement: Models hosted and trained abroad serve domestic minors with no local establishment to proceed against. e.g. Indian users accessing chatbots operated entirely from other jurisdictions.
    5. Absence of a dedicated statute: Regulation runs on instruments written for hosted content and for data, not for generated responses. e.g. India relying on the Information Technology Act, 2000 and platform led interventions.
    6. Parental digital literacy gap: Controls assume a supervising adult who understands the product. e.g. first generation smartphone households where the child is the more capable user.
    7. Design based harm outside content rules: Engagement optimisation and empathy simulation are not content categories that a takedown regime can reach. e.g. China moving to restrict AI systems that encourage emotional dependence, a design level rather than content level intervention.

    Way Forward

    1. Independent safety testing: Require third party red team testing of chatbot behaviour with adolescent personas, with results published rather than held by the developer.
    2. Statutory age assurance standards: Prescribe a privacy preserving age assurance method so protection does not depend on a company’s own inference or on a child’s self declaration.
    3. Default on, not opt in: Make the safest configuration the default for accounts assessed as belonging to minors, so protection does not depend on a parent activating it.
    4. Duty of care by design: Place an explicit obligation on developers to design against engagement maximisation and emotional dependence for minors, following the design level approach rather than a content list.
    5. Crisis routing obligations: Mandate that any self harm, suicide or eating disorder cue in a minor’s conversation trigger an immediate handoff to a human helpline, with logged compliance.
    6. Transparency reporting: Require periodic public reporting of safety failure rates, bypass incidents and the duration effect on guardrail reliability in long conversations.
    7. Digital literacy and sensitisation: Build chatbot specific awareness into school curricula and parent outreach, since the risk is a design property that neither group currently recognises.
    8. A dedicated Indian instrument: Move from the present patchwork to a clear framework for artificial intelligence products used by minors, backed by the National Commission for Protection of Child Rights and the data protection regulator.

    “[2025, GS2, 15 marks] The National Commission for Protection of Child Rights has to address the challenges faced by children in the digital era. Examine the existing policies and suggest measures the Commission can initiate to tackle the issue.”

  • To build AI for all, bring in more women

    Why in the News

    India ranks among the world’s leading artificial intelligence ready nations, powered by Digital Public Infrastructure and a large innovation ecosystem, while women fall from 43 percent of STEM graduates to 10 percent of senior AI leadership. Every artificial intelligence system begins with data and every dataset begins with people, so a pipeline that loses women at each stage produces systems that reproduce the inequality of the society they learn from.

    What is the AI pipeline?

    1. Definition: The AI pipeline is the full sequence from data collection through model training and deployment to the decisions the model produces.
    2. Not only technical: It is not merely a technological conduit of code, silicon and compute power. It is fundamentally a human pipeline.
    3. It starts early: The pipeline begins before the first line of code is written, at the point where data about people is collected or not collected.
    4. Where the consequences land: Its outputs shape decisions affecting millions, from loan sanction to clinical recommendation.
    5. The failure mode: When people are absent from that data, artificial intelligence inherits those gaps.

    What is Digital Public Infrastructure?

    1. Definition: Digital Public Infrastructure (DPI) is a set of shared, interoperable digital systems, such as digital identity, payments and data exchange layers, built as public utilities on which both government and private services run.
    2. Why it matters here: India’s artificial intelligence readiness is powered by DPI, which also determines whose transactions and records enter the datasets models are trained on.

    What is the India AI Mission?

    1. Definition: The India AI Mission is the national programme providing compute capacity, datasets, application development support, skilling and startup financing for artificial intelligence in India.
    2. Relevance here: It is the vehicle through which artificial intelligence in India can be steered onto the same inclusive path that DPI followed for public welfare.

    Where does the pipeline leak women?

    1. STEM foundation: Women account for 43 percent of India’s STEM graduates, one of the world’s largest pools of women STEM graduates.
    2. Tech workforce: Representation falls to 26 percent in the technology workforce.
    3. Advanced AI roles: Only 12 percent of professionals in advanced artificial intelligence roles are women.
    4. Senior AI leadership: Women hold just 10 percent of senior artificial intelligence leadership positions.
    5. What the sequence shows: At every stage the pipeline leaks talent, lived experience and innovation, so the loss compounds rather than occurring at one bottleneck.

    What causes the leakage?

    1. Access to the network itself: Only 57 percent of women have independent internet access, compared with 72 percent of men.
    2. Nutrition and education: Unequal nutrition and unequal education set the disparity before any career choice is made.
    3. Caregiving responsibilities: Unpaid care work removes women from the workforce at the point where advanced technical careers compound.
    4. Workplace discrimination: Discrimination at work blocks progression from entry level technical roles into advanced ones.
    5. Language barriers: Artificial intelligence education is dominated by English, which excludes those schooled in other languages.
    6. School infrastructure: A student cannot pursue robotics where her school lacks the necessary infrastructure, so the exclusion begins well before higher education.
    7. Influence, not only presence: A woman who becomes an artificial intelligence engineer often remains the only woman in the room, with limited influence in product design.

    What happens to systems built without women in the data?

    1. Credit assessment: A self help group member in rural Bihar applying for a micro-loan is scored by models relying mainly on historical male financial patterns, which may underestimate her creditworthiness.
    2. Maternal health tools: A community health worker in Gujarat depends on artificial intelligence enabled maternal health tools, and training data that fails to reflect local nutrition and health conditions produces inaccurate recommendations affecting maternal care.
    3. The general mechanism: Artificial intelligence automates existing inequalities when trained on incomplete or biased data.
    4. The learning relationship: Artificial intelligence learns from society, so an unequal society produces an artificial intelligence that reflects that inequality.
    5. Why datasets alone are insufficient: Correcting the output requires more than diverse datasets, because the decisions about what to collect and what to optimise are made by the people in the room.

    Does India’s AI readiness conceal an exclusion problem?

    1. The readiness claim: India ranks among the world’s leading artificial intelligence ready nations, powered by Digital Public Infrastructure and a thriving innovation ecosystem.
    2. The contradiction beneath it: India produces one of the world’s largest pools of women STEM graduates, and women steadily disappear as the artificial intelligence pipeline advances.
    3. Formal equality achieved early: When India adopted its Constitution in 1950, it granted women and men universal adult franchise simultaneously, ahead of the sequence followed in several western democracies.
    4. Substantive access lagging: That simultaneous political inclusion sits alongside a 15 percentage point gap in independent internet access between men and women today.
    5. What the measure of leadership should be: True artificial intelligence leadership cannot be measured only by models, investments or patents. It must be measured by whether artificial intelligence reflects India’s diversity of languages, cultures, socio-economic realities and lived experiences.

    What does the corrective path look like?

    1. The precedent of scale: India has already shown how technology can advance public welfare at scale, and the India AI Mission offers the opportunity to ensure artificial intelligence follows the same inclusive path.
    2. Existing women’s institutions: Across rural India, women’s self-help groups have built strong financial ecosystems through collective savings and entrepreneurship, which is usable financial data and an existing delivery network.
    3. Influence changes output: When women occupy positions of influence, the technology itself shifts.
    4. Four roles, not one: Women and marginalised communities must participate as researchers, engineers, entrepreneurs and policymakers, not only as subjects in the training data.
    5. The constitutional foundation: The commitment to simultaneous inclusion continues through Digital Public Infrastructure, which provides the base for building inclusive artificial intelligence.

    Challenges to building inclusive AI

    1. Unpaid care work truncates technical careers: Time available for advanced training and long project cycles is unequal, e.g. the Time Use Survey 2019 recorded women spending 299 minutes a day on unpaid domestic work against 97 minutes for men.
    2. Device and connectivity gap precedes the skills gap: Independent access, not shared household access, determines who generates data, e.g. the National Family Health Survey 2019 to 2021 found 33.3 percent of women had ever used the internet against 57.1 percent of men.
    3. Language exclusion in model and curriculum: English dominant material and models exclude most first generation learners, e.g. Bhashini and BharatGen were set up precisely because Indian language coverage in large models was thin.
    4. Data annotation labour has no design voice: The workers who label training data are outside the decisions the data shapes, e.g. annotation work is outsourced at low wages with no representation in product design.
    5. No bias audit obligation: Automated decision systems face no statutory fairness testing requirement, e.g. the Digital Personal Data Protection Act, 2023 governs consent and processing of personal data but imposes no algorithmic audit duty.
    6. Online safety drives women off the platforms that generate data: Harassment reduces sustained participation, e.g. National Crime Records Bureau data has recorded a rising count of cyber crimes against women.
    7. Absence of sex disaggregated public datasets: Models cannot be checked for differential performance where the data does not record the split, e.g. many administrative datasets used for training carry no reliable gender field.

    Conclusion

    The central point is that the artificial intelligence pipeline is a human pipeline, and the numbers show it losing women at every stage from 43 percent of STEM graduates to 10 percent of senior AI leadership. Diverse datasets alone will not correct outputs shaped by rooms in which women are absent, so participation must extend to research, engineering, entrepreneurship and policymaking. What remains unresolved is the access gap that precedes all of it, with only 57 percent of women holding independent internet access against 72 percent of men.

  • Gene Editing’s Bold Move: Permanently Shut Down PCSK9

    Why in the News

    VERVE-102, an experimental in vivo base editing therapy delivered as a single intravenous infusion, permanently switches off the PCSK9 gene inside liver cells and cut LDL cholesterol by about 62 percent in a phase 1 trial. Cholesterol control has until now been a lifelong compliance problem, and a one time genetic change replaces that problem with a permanent, irreversible one.

    How does VERVE-102 work?

    1. What it is: VERVE-102 is not a traditional drug. It is a form of in vivo gene editing, meaning the editing is done inside the patient’s body rather than on cells removed and returned.
    2. Step 1, delivery: Genetic instructions are delivered through a single intravenous infusion.
    3. Step 2, the edit: Those instructions make a one time targeted change to the DNA inside liver cells, altering a single base in the PCSK9 gene.
    4. Step 3, the effect: The edited liver cells permanently lose the ability to produce PCSK9.
    5. Step 4, the outcome: With PCSK9 production switched off, the liver clears more LDL cholesterol from the blood, and the effect persists without repeat dosing.
    6. The stated goal: A single infusion that permanently reduces the liver’s ability to produce PCSK9, so that a one and done cholesterol treatment could eventually replace conventional medicines.

    What is LDL cholesterol?

    1. Definition: LDL (low-density lipoprotein) is called bad cholesterol because high levels make it stick to artery walls and form hard fatty deposits called plaque.
    2. Why it matters: These deposits narrow the arteries and block blood flow, which raises the risk of heart attacks and strokes.

    What is PCSK9 and why is it the target?

    1. What it is: PCSK9 is a protein involved in regulating LDL cholesterol in the blood.
    2. The natural experiment: People who naturally carry certain loss-of-function changes in the PCSK9 gene have lower LDL cholesterol throughout their lives and a lower risk of coronary heart disease.
    3. The inference: Reducing PCSK9 activity is therefore a safe and effective route to lowering cardiovascular risk.
    4. Confirmed by drugs: PCSK9 monoclonal antibodies substantially reduce LDL cholesterol and cardiovascular events, confirming the target.
    5. The limitation VERVE-102 addresses: Traditional medicines temporarily block PCSK9 or reduce its production, so their effects require continued treatment.

    What did the phase 1 trial find?

    1. LDL reduction: LDL cholesterol fell by about 62 percent in the highest dose group after four weeks.
    2. PCSK9 reduction: PCSK9 levels in that group fell by about 88 percent.
    3. Absolute fall: LDL cholesterol decreased by approximately 78 mg/dL on average.
    4. Follow up length: Some participants were followed for at least one year, and the longest follow up reached 18 months.
    5. Durability so far: The reductions in PCSK9 and LDL cholesterol were relatively stable across that period.

    How much cardiovascular risk does that reduction translate into?

    1. The established ratio: For every 1 mmol/L reduction in LDL cholesterol, cardiovascular risk falls by 20 to 22 percent.
    2. Worked case: An LDL cholesterol of 4.0 mmol/L, approximately 155 mg/dL, falling to 1.6 mmol/L is a 60 percent reduction.
    3. Effect of that case: That fall halves the patient’s cardiovascular risk.
    4. What remains unproven: VERVE-102 has not yet been shown to prevent heart attacks or strokes directly.
    5. The supporting evidence: All cholesterol lowering trials so far have shown that lower cholesterol means fewer cardiovascular events, and drugs blocking the PCSK9 protein have been shown to reduce heart attacks.

    How does it compare with the treatments already in use?

    1. Statins: Usually the foundation of treatment. They are relatively inexpensive, widely available, and supported by extensive evidence showing reductions in cardiovascular events.
    2. Ezetimibe: A cholesterol absorption inhibitor, taken orally, that works by blocking cholesterol from being absorbed in the small intestine.
    3. PCSK9 antibody medicines: They produce powerful LDL reductions and have demonstrated cardiovascular benefits, but require repeated injections.
    4. Inclisiran: It reduces PCSK9 production and can lower LDL cholesterol by roughly 50 percent, with less frequent dosing that makes long term treatment easier. It does not permanently modify DNA.
    5. The distinguishing feature of VERVE-102: Every existing option acts temporarily and must be continued. VERVE-102 makes a permanent change to DNA.

    Does permanence justify the loss of reversibility?

    1. The compliance case: Repeat prescriptions and remembering daily doses are a standing burden, and a safe one time treatment would remove that burden entirely.
    2. The unknown: This is a permanent change and the long term consequences are not yet known, so treated patients will need close observation.
    3. The reassurance from biology: Naturally occurring loss-of-function mutations of the gene exist, and people carrying them have less heart disease and live longer, which is the basis for the trial.
    4. The evidence horizon problem: An 18 month period is very different from proving that an effect will last for decades, and that requires further research.
    5. The current standing of the therapy: It is a potential future option for selected high risk patients, not a replacement for statins, ezetimibe, PCSK9 inhibitors or inclisiran.
    6. Trial breadth: More diverse trials are needed to establish whether the effect holds across populations over decades.

    Who would be considered for it first?

    1. Familial hypercholesterolemia: An inherited condition producing very high LDL cholesterol from birth, whose patients have the most to gain from a permanent reduction.
    2. Very high cardiovascular risk patients: Those whose risk is not controlled by existing therapy would be the second group.
    3. The staging logic: Beginning with these groups allows observation for problems before any wider use.
    4. What it is not yet: It is not a population level cholesterol intervention and is not positioned as one.

    Challenges to VERVE-102

    1. Irreversibility of a permanent edit: A therapy that cannot be stopped removes the physician’s ability to withdraw treatment, e.g. a statin prescription can be discontinued the day an adverse effect appears, while an edited liver cell population cannot be restored.
    2. Evidence horizon is short: Durability is established only to 18 months, e.g. statin cardiovascular outcome evidence rests on trials such as the Heart Protection Study that ran over five years in more than 20,000 participants.
    3. Delivery vector and off target risk: Gene therapy delivery carries historical safety precedent, e.g. the 1999 death of a participant in an adenoviral vector gene therapy trial in the United States halted the field for years.
    4. Cost and access: One time genetic therapies have been priced far beyond public health budgets, e.g. Casgevy, the first approved CRISPR based therapy, is priced at over two million dollars per patient in the United States.
    5. Population applicability: Early phase cohorts do not establish effect across differing lipid profiles, e.g. coronary artery disease in South Asians presents roughly a decade earlier and at lower body mass index than in western populations.
    6. Regulatory pathway for permanent somatic edits: Approval frameworks for irreversible somatic edits are still forming, e.g. India’s National Guidelines for Gene Therapy Product Development and Clinical Trials, 2019 permit somatic editing under review but bar germline editing outright.
    7. The competing benchmark is already cheap: A one time therapy must justify a large upfront price against an existing generic, e.g. statins cost a few rupees a day in India and are on the National List of Essential Medicines.

    Conclusion

    The central finding is that a permanent genetic switch off of PCSK9 through a single infusion produces LDL reductions larger than any daily medicine achieves, and that the reduction has held for 18 months. What remains unresolved is whether a permanent change is safe across a lifetime, and whether the LDL reduction converts into fewer heart attacks and strokes, neither of which the phase 1 data can answer. Until large outcome trials report, the therapy stands as an option for familial hypercholesterolemia and very high risk patients rather than a replacement for statins, ezetimibe, PCSK9 inhibitors or inclisiran.

    PYQ Relevance:

    Question (2021, GS3): “What are the research and developmental achievements in applied biotechnology? How will these achievements help to uplift the poorer sections of society?
    Linkage: Applied biotechnology is the primary field where gene editing techniques (like CRISPR) are developed to address challenges in health and agriculture, which can specifically benefit the underprivileged

  • [18th August 2026] The Hindu OpED: Match AI models to workloads, not leaderboards

    PYQ Relevance
    Question (2024, GS4): “The application of Artificial Intelligence as a dependable source of input for administrative rational decision-making is a debatable issue. Critically examine the statement from the ethical point of view”
    Linkage: Administrative tasks require balancing capability with governance. The article  argue that leaderboards measure capability on standard tasks but fail to predict production quality or address the ethical/safety guardrails needed for specific organizational workloads

    Why in the News

    A new artificial intelligence (AI) release claims the top of some leaderboard almost every week, and enterprises that once simply consumed the strongest available model through a managed interface now face a harder choice. What determines success is no longer which model scores highest but which model and which deployment approach fit a particular workload, with cost, governance, data residency and intellectual property protection now sitting alongside raw capability. A security incident in July 2026 made the point concrete, when a frontier model’s own safety controls blocked the forensic work and the investigation had to be completed on a self hosted model.

    What are open weight models?

    1. What they are: Models whose trained weights are released so that an organisation can download and run them on its own infrastructure, subject to the licence terms.
    2. How they differ from closed models: A closed model is delivered as a remote service, and the organisation never holds the parameters that do the computation.
    3. The data effect: Sensitive data can remain inside approved environments rather than being transmitted to an external provider.
    4. The customisation effect: Models can be fine tuned on proprietary knowledge without routinely sending that knowledge to an external provider.
    5. The commercial effect: Enterprises gain greater portability, reduce dependence on any single vendor’s road map and pricing, and often see substantially lower per token costs.
    6. The important qualification: Total cost of ownership still depends heavily on utilisation and scale, so the lower unit price does not automatically mean a lower bill.

    What is a frontier model?

    1. What it is: The most capable general purpose model a leading laboratory currently offers, delivered as a remote service through a commercial interface.
    2. Where it fits: Customer facing tasks that demand the highest reasoning capability often belong on these closed services.

    What is data residency?

    1. What it is: A requirement that data be stored and processed within a specified national or legal jurisdiction.
    2. Why it drives deployment choice: A regulated workload subject to a residency obligation cannot be served by a model hosted outside that jurisdiction, whatever its benchmark score.

    What is token sovereignty?

    1. What it is: The objective of having artificial intelligence computation for a country’s users performed on infrastructure located and governed within that country.
    2. What the term refers to: A token is the unit in which model input and output are measured and billed, so sovereignty over tokens means sovereignty over where inference actually runs.

    What is managed inference?

    1. What it is: A service that hosts open weight models on controlled infrastructure and exposes them to customers through managed endpoints.
    2. What it removes: The customer gets data residency and fine tuning flexibility without having to build and operate the underlying graphics processing unit clusters and the inference serving stack.

    What is fine tuning?

    1. What it is: Further training of an already trained model on an organisation’s own data so that it performs better on that organisation’s specific tasks.
    2. Why it raises a control question: Fine tuning on proprietary knowledge means that knowledge must be exposed to whoever controls the training environment.

    What are safety guardrails?

    1. What they are: Controls built into a model service that refuse categories of request judged harmful, applied before the model responds.
    2. Their structural limitation: They operate on the content of the request, so they cannot distinguish an authorised security responder from an attacker submitting the same material.

    Why has model ranking stopped being the deciding factor?

    1. The churn problem: A new release claims the top of some leaderboard almost every week, so a ranking based decision is obsolete within weeks.
    2. The old default: Until recently most enterprises simply chose the strongest available model and consumed it through managed interfaces from the frontier laboratories.
    3. What now sits alongside capability: Cost, governance, data residency, intellectual property protection and operational complexity are now first order considerations, not secondary ones.
    4. The reframed question: The question is not which model scores highest but which model and which deployment approach are right for a particular workload.
    5. What a benchmark cannot capture: A leaderboard measures capability on a standard task set and says nothing about where the data goes or what the workload costs at production volume.
    6. The decision level: The call belongs at the level of the individual workload rather than at the level of a single corporate standard.

    What did the July 2026 security incident demonstrate?

    1. The trigger: An AI driven intrusion hit the infrastructure of a major model hosting company in July 2026.
    2. The first response: Incident responders first turned to frontier models behind commercial interfaces to analyse thousands of attacker actions.
    3. What the forensic work required: Feeding real exploit payloads, attack logs and command and control artifacts to the models.
    4. What blocked it: The providers’ safety guardrails blocked the requests, because the systems could not distinguish an authorised responder from an attacker.
    5. How it was resolved: The company completed the analysis on a self hosted open weight model instead.
    6. The data consequence: Sensitive incident data stayed inside its own environment throughout that analysis.
    7. The correct reading: The lesson was not that closed models are inferior, it was that some workloads structurally require a model the organisation controls.
    8. The class of affected work: Security forensics, malware analysis and any investigation that must examine genuine attacker tooling cannot tolerate third party guardrails that refuse the query.
    9. The preparedness point: A capable, vetted open weight model must already be running on infrastructure the organisation governs before an incident occurs, not after.

    Why can one deployment strategy not serve every workload?

    1. The basic fact: Very few organisations have only one artificial intelligence workload.
    2. Banking against marketing: A bank analysing confidential customer data has different requirements from a marketing team generating campaign content.
    3. Manufacturing against cyber security: A manufacturer embedding AI in customer service has different priorities from a cyber security team examining malware.
    4. The control axis: Enterprises must classify workloads by control requirements as rigorously as by performance needs.
    5. What the classification decides: The control requirement, not the capability score, is what determines whether a workload can sit on a remote service at all.
    6. The realism check: Expecting one model and one deployment strategy to fit every use case is increasingly unrealistic.

    Why are open weights not a free option?

    1. The easy part: Downloading a model is the easy part of the exercise.
    2. What operation actually needs: Running it reliably at enterprise scale requires graphics processing unit infrastructure, inference serving, monitoring, security, governance, upgrades and licensing.
    3. The trade stated plainly: Greater control comes with greater responsibility.
    4. Where the trade works: For large organisations with deep engineering capacity the trade off can be worthwhile.
    5. Where it does not: For most mid sized and small enterprises it is far more challenging.
    6. The cost qualification: Lower per token cost does not settle the question, because total cost of ownership depends on utilisation and scale.

    What is the third deployment option now emerging?

    1. What it is: Managed inference platforms for open weight models, which host leading open weight families on controlled infrastructure and expose them through managed endpoints.
    2. What the enterprise gets: Many of the benefits of open weights, namely data residency, fine tuning flexibility and often lower cost.
    3. What the enterprise avoids: Building and operating the underlying graphics processing unit clusters and the inference stack.
    4. The Indian example: Sarvam Inference, an India hosted managed service unveiled at a 2026 conference, is one concrete instance of the category taking shape.
    5. What it serves: The platform currently serves a 105 billion parameter domestic model alongside leading open weight families such as GLM 5.2 and Gemma 4, all running on domestic infrastructure.
    6. Where the significance lies: The significance is not any individual model, since enterprises could already download many of them.
    7. The actual problem solved: The challenge was making them work reliably in production, which means handling concurrency, latency, security and continuous updates at scale.
    8. The access effect: Production grade endpoints under Indian data residency are likely to democratise access for companies that could never justify specialised AI operations teams.
    9. The policy effect: It supports the broader push for token sovereignty.

    Where does the case for control run into its own limit?

    1. The caveat stated: Managed open weight platforms reintroduce vendor dependence.
    2. Where the dependence moves to: It shifts from the model layer to the infrastructure layer, and it does not disappear.
    3. What must therefore be tested: Enterprises should evaluate portability guarantees, security posture, pricing trajectory and exit paths.
    4. The standard to apply: The same rigour applied to any frontier interface contract must be applied to the managed open weight provider.
    5. Why this is the real tension: The reason to leave a closed provider was concentration risk, and the managed route recreates that risk one layer down.
    6. What it does not undo: Data residency and the ability to run forensic workloads are genuinely gained, so the answer is a different contract, not a return to the closed default.

    What do sovereign artificial intelligence efforts elsewhere show?

    1. European Union: The AI Act, adopted in 2024, is the first comprehensive horizontal law on artificial intelligence, and it classifies systems by risk tier with obligations attached to each.
    2. European Union infrastructure: The GAIA-X initiative was created to build a federated European cloud and data infrastructure with defined residency and portability rules.
    3. France: A domestic laboratory has built and released open weight model families, which is the European route to reducing dependence on United States providers.
    4. United Arab Emirates: The Falcon open weight model family was released by a state backed research institute as a deliberate sovereign capability investment.
    5. China: Several Chinese laboratories release strong open weight models, and the GLM family named in this discussion is one of them, which is how open weights have become geopolitically distributed rather than concentrated.
    6. Japan and South Korea: Both have funded national language model programmes on domestic compute, on the same reasoning of language coverage and residency.
    7. What the pattern demonstrates: Sovereignty efforts everywhere target the infrastructure and weights layer rather than benchmark leadership, which is the same shift the enterprise level argument describes.

    How should a workload be matched to a deployment model?

    1. Customer facing reasoning tasks: Tasks demanding frontier reasoning often fit closed interfaces from the leading laboratories.
    2. Regulated workloads: Workloads with strict data residency obligations frequently suit managed open weight platforms hosted in country.
    3. Security and intellectual property work: Security forensics, malware analysis and intellectual property critical fine tuning usually belong on self hosted deployments.
    4. The discipline required: The call must be made workload by workload rather than by corporate default.
    5. What the organisation must understand: The strengths, limitations and economics of each approach, so the match is made on evidence rather than on habit.
    6. The balance being struck: Every workload should go to the option delivering the right balance of capability, control, cost and governance.
    7. The organisational conclusion: Deployment choice is a core architectural decision, not a procurement afterthought.

    Challenges to workload based artificial intelligence deployment

    1. Absence of a workload classification discipline: Most enterprises have no register of which workloads carry control obligations, so the match cannot be made. e.g. regulated entities discovering only during an audit that customer data was processed through an overseas endpoint.
    2. Graphics processing unit scarcity and cost: Self hosting requires accelerator capacity that is expensive and supply constrained. e.g. the IndiaAI Mission’s empanelment of compute providers to make subsidised graphics processing units available because market capacity was insufficient.
    3. Licence ambiguity in open weights: Open weight licences often restrict commercial use or downstream redistribution, which is discovered late. e.g. community licences that cap monthly active users or bar use in training competing models.
    4. Guardrail rigidity in legitimate work: Safety controls block authorised security and medical work because they judge content, not authorisation. e.g. the July 2026 forensic analysis that had to be moved to a self hosted model.
    5. Skills concentration: Inference serving, quantisation and model operations skills sit in a small number of firms. e.g. mid sized enterprises unable to staff a dedicated AI operations team and therefore defaulting to a single vendor.
    6. Model supply chain risk: Downloaded weights and their dependencies can carry tampered artifacts. e.g. malicious serialised model files uploaded to public model hubs and later removed.
    7. Evaluation gap: Public benchmarks do not measure performance on an enterprise’s own tasks, so a leaderboard rank does not predict production quality. e.g. contamination of benchmark test sets in model training data inflating reported scores.
    8. Cross border transfer restrictions: Data protection law limits where personal data may be processed, which constrains model choice. e.g. restrictions on transfer of personal data to notified countries under India’s data protection statute.
    9. Vendor lock in at the infrastructure layer: A managed provider’s proprietary serving stack and pricing can be as sticky as a closed model contract. e.g. fine tuned model artefacts that cannot be exported and rehosted elsewhere.

    Way Forward

    • Invest in AI skills and secure open-weight ecosystems covering inference serving, model evaluation, quantisation, monitoring and supply-chain security.
    • Adopt workload-based AI deployment by matching each use case with the right balance of capability, cost, control and governance.
    • Build domestic AI infrastructure including GPU capacity, managed inference platforms and secure data centres to strengthen token sovereignty.
    • Strengthen AI governance through clear workload classification, data residency rules, licensing checks and security standards.
    • Develop hybrid and portable architectures to avoid dependence on a single model or infrastructure provider, with clear exit and portability provisions.

  • NASA’s Moon Base: What India will gain by joining

    Why in the News

    The National Aeronautics and Space Administration (NASA) has invited the Indian Space Research Organisation (ISRO) to join its Moon Base programme, a permanent crewed research station to be built on the Moon in stages. The invitation forces a choice between building an independent human spaceflight, space station and lunar landing capability at national cost, and acquiring the same capability faster inside a programme the United States leads. India signed the Artemis Accords in 2023 as the 27th nation, so the diplomatic ground for joining is already laid.

    What is the NASA Moon Base programme?

    1. What it is: A permanent research station on the lunar surface that astronauts and robots can inhabit for prolonged periods.
    2. What it is for: It is meant to facilitate research and to allow exploration and exploitation of lunar resources.
    3. How it is built: The base is assembled in stages over several years, requiring repeated crewed and robotic trips to the Moon.
    4. Its scale: In scale and ambition it compares only with the Apollo missions, and it could be the costliest scientific project ever undertaken.
    5. Its engineering claim: It will possibly be the most challenging engineering exercise ever attempted by humanity.
    6. Its delivery model: NASA will not execute it alone and is seeking partners in both the international community and private industry.

    What is the Bharat Antariksh Station?

    1. What it is: India’s planned indigenous space station, to be built and operated by ISRO as a crewed orbital facility.
    2. Why it is cited here: ISRO must hold the technology to build such infrastructure, and India is unlikely within about a decade to have a scientific ecosystem needing an entire station for its own use all year round.

    What is the lunar South Pole?

    1. What it is: The polar region of the Moon holding permanently shadowed craters where water ice is expected to survive.
    2. Why it is the target: Phase One of the Moon Base programme sends robotic missions specifically to the South Pole, because water ice can be converted into drinking water, breathable oxygen and rocket propellant.

    What are interoperable systems?

    1. What they are: Common standards and hardware interfaces that let equipment built by different countries connect and work together in space.
    2. What the Accords require: Signatories emphasise interoperability in fuel storage, landing structures, communications systems and power systems, which is a light obligation for ISRO because it is only beginning to develop these systems.

    What is deglobalisation?

    1. What it is: The retreat from shared international supply chains towards national self reliance in a strategic technology.
    2. Where it currently applies: Semiconductors, clean energy and artificial intelligence, where supply chains and resources are controlled by a small set of actors.

    What are the three phases of the Moon Base programme?

    1. Phase One, now to 2029: Focus on gaining reliable access to the lunar surface and building a deeper understanding of the environment.
    2. Phase One activity: Robotic missions will explore the lunar South Pole, demonstrate new technologies and gather the knowledge needed to guide future development.
    3. Phase Two, 2029 to 2032: NASA will begin deploying the first infrastructure needed to support long term operations on the Moon.
    4. Phase Two systems: Early power systems, cargo transportation, logistics and communications capabilities will expand the human footprint and enable increasingly complex missions.
    5. Phase Three, 2032 and beyond: NASA will begin assembling a permanent lunar outpost where astronauts can live and work for extended periods.
    6. Phase Three systems: Habitats, power systems, communications, transportation and other critical capabilities will support an enduring human presence.

    Why is NASA seeking partners instead of building the base alone?

    1. Budget compression: NASA’s budget has been cut significantly under the current US administration, so a solo build is not affordable.
    2. Shift of manufacturing: Most of NASA’s hardware production now happens in the private sector rather than in house.
    3. Two partner pools: It is seeking collaboration both from the international community and from private industry.
    4. A ready pool of states: The 70 countries that signed the Artemis Accords have already signalled a willingness to join such a collaboration.
    5. Cost of the mission profile: Repeated crewed and robotic trips to the Moon over several years put the cost beyond a single agency’s programme line.
    6. Precedent: The International Space Station established that a permanent crewed facility is built and run as shared infrastructure, not as one nation’s asset.

    Why can ISRO not sustain its lunar and station ambitions on its own?

    1. Three simultaneous programmes: ISRO is running an independent human spaceflight programme, a Moon landing programme and a full fledged space station programme in parallel.
    2. Capability against sustainability: Holding these capabilities is important, and running them sustainably on India’s own scientific and economic base is a separate question.
    3. The demand problem: India is unlikely, within about a decade, to have a scientific ecosystem hungry enough to occupy an entire space station all year round.
    4. The cost of lunar exploration: A separate full fledged lunar exploration programme carries costs that are prohibitive even for the world’s richest economy.
    5. Competing national goals: India is chasing multiple parallel goals on the path to prosperity, which limits how much can be allocated to space at the scale required.
    6. The shared infrastructure conclusion: The Bharat Antariksh Station will have to be shared infrastructure on the model of the International Space Station.

    What does ISRO gain by joining the Moon Base programme?

    1. Mission management experience: Participation gives ISRO experience in planning and executing complex missions of exactly the type it intends to run itself.
    2. Technology leapfrog: It allows ISRO to skip development stages rather than rebuild capability that already exists elsewhere.
    3. The obsolescence clock: Space exploration has reached a stage where a 10 year gap in technology development can leave a nation well behind.
    4. Avoiding duplication: There is no economic sense in reinventing capability that a partner already holds.
    5. Timeline compression: Cooperation with the United States lets ISRO fast track its own project timelines and reach the frontiers of technology development.
    6. Spin off benefits: The collaboration can generate spin off technologies with cascading dividends across sectors beyond space.
    7. No exclusivity cost: Signing the Accords or joining the Moon Base effort does not prevent India from continuing its long standing space cooperation with Russia.

    What do the positions of other space powers reveal about the Accords?

    1. Japan: A signatory and a major space faring nation, contributing habitation and pressurised rover work to the Artemis effort.
    2. South Korea: A signatory that has built an independent lunar capability, having placed the Danuri orbiter around the Moon in 2022.
    3. Israel: A signatory whose Beresheet lander attempt in 2019 made it one of the few states to have reached lunar orbit.
    4. European states: Several are signatories, and the European Space Agency separately supplies the service module for NASA’s Orion crew vehicle.
    5. Russia: Not a signatory, and it is attempting a comparable lunar effort through its own partnership.
    6. China: Not a signatory, and it is pursuing the same objective with Russia rather than through the Accords.
    7. What the split demonstrates: The absence of the two other major space powers is what gives the criticism of a US led camp its credibility.

    Are the Artemis Accords a US led bloc that bypasses multilateral arrangements?

    1. The formal position: The Accords are a set of principles and good practices that countries agree to follow in their space activities.
    2. The criticism: They are increasingly seen as a US led camp writing its own rules for space exploration and the use of extraterrestrial resources, subtly bypassing international multilateral arrangements.
    3. What lends the criticism weight: Russia and China, the two other major space powers, are outside the grouping.
    4. India’s historical reluctance: India has traditionally been extremely reluctant to join any such grouping, and it still joined as an early signatory, the 27th nation, in 2023.
    5. The counter argument, no exclusion: Space is not adversarial at present, and a US landing on the Moon does not give it control over the area or its resources.
    6. The counter argument, no scarcity: The Moon is large enough and its resources abundant enough to support the efforts of all parties in the foreseeable future.
    7. The counter argument, no supply chain lock: There is no domination of supply chains or control over resources in space, so the deglobalisation trend seen in semiconductors, clean energy and artificial intelligence does not transfer to this case.
    8. The residual risk: The real exposure is technological, not geopolitical, and it is the possibility of ISRO getting locked into the US technology ecosystem to the extent of overdependence.

    Challenges to India joining the Moon Base programme

    1. Technology ecosystem overdependence: Deep integration with one partner’s standards makes later substitution expensive. e.g. India’s dependence on Russian cryogenic engine technology in the 1990s stalled the GSLV programme for over a decade after the Missile Technology Control Regime pressure on the transfer.
    2. Programme discipline slipping: Collaboration can crowd out ISRO’s own milestones if targets are not separately protected. e.g. the Gaganyaan crewed flight has already moved from its original 2022 target to the later part of this decade.
    3. Export control friction: Dual use hardware transfers remain governed by US licensing that can be withheld. e.g. International Traffic in Arms Regulations clearances have historically delayed satellite component supplies to Indian entities.
    4. Budget asymmetry: India’s civil space spending is a small fraction of NASA’s, which limits its bargaining position on workshare. e.g. the Department of Space’s annual budget is of the order of Rs 13,000 crore against a NASA budget many times larger.
    5. Launch reliability: A partner role demands schedule certainty that India’s recent launch record does not yet demonstrate. e.g. three of the six ISRO missions in 2025 and 2026 failed to place satellites in the intended orbits.
    6. Balancing the Russia relationship: Deeper alignment with the Accords sits alongside a long standing space partnership that must be maintained separately. e.g. Russian support for the crew module and life support work under the Gaganyaan programme, including astronaut training at the Gagarin Cosmonaut Training Centre.
    7. Unsettled resource law: The Accords permit extraction and use of space resources, and that reading of the Outer Space Treaty is contested. e.g. the Moon Agreement of 1979 treats lunar resources as the common heritage of mankind and has been ratified by very few states.
    8. Volatile bilateral politics: The India United States relationship has been unstable in the last two years, which is a risk for a multi decade commitment. e.g. trade and tariff disputes running alongside this civil space engagement.

    Conclusion

    The Moon Base invitation converts an abstract question about strategic autonomy into a concrete question about economic sustainability. India can build the technology for a station and a lunar landing on its own, and it is unlikely to be able to run either sustainably at national scale, which is why joining offers a genuine leapfrog rather than a compromise. The condition that must hold is that ISRO protects its own targets and avoids locking itself into a single technology ecosystem while it collaborates.

    “[2025] Consider the following space missions:
    I. Axiom-4
    II. SpaDeX
    III. Gaganyaan
    How many of the space missions given above encourage and support microgravity research?
    (a) Only one
    (b) Only two
    (c) All the three
    (d) None

  • GISAT-1A take-off in September to end ISRO’s seven-month operational hiatus

    Why in the News

    The Indian Space Research Organisation (ISRO) is set to resume launches in the first week of September 2026 with GISAT 1A, after a seven month operational pause.

    The pause followed multiple mission failures and has affected NavIC, which currently has only 3 operational satellites, against the 4 required for basic standalone positioning.

    What is GISAT 1A?

    • GISAT: Geo Imaging Satellite
    • Also designated EOS 05.
    • Earth observation satellite with a 10-year mission life.
    • Provides frequent imaging of large areas.
    • Applications include disaster monitoring, agriculture and forestry.
    • It replaces GISAT 1 / EOS 03, which failed to reach orbit in 2021.

    What is NavIC?

    • NavIC: Navigation with Indian Constellation
    • Formerly called IRNSS: Indian Regional Navigation Satellite System.
      • Developed by ISRO.
      • Provides Positioning, Navigation and Timing (PNT) services.
      • Covers India and surrounding regions.
      • Reduces dependence on foreign navigation systems.
      • Currently operational: IRNSS 1B, IRNSS 1I and NVS 01.

    Why are 4 Satellites Needed?

    • Positioning requires signals from at least 4 satellites to determine:
      • Three-dimensional position
      • Receiver clock error
    • With only 3 satellites, NavIC cannot provide standalone positioning, though its timing service remains functional.

    What is PNT?

    • Positioning: Determines location.
    • Navigation: Determines movement and route.
    • Timing: Provides precise time reference.

    Why Did ISRO’s Launch Calendar Stall?

    Three of six missions during 2025 and 2026 failed to achieve their intended objectives:

    • PSLV C61 / EOS 9: Third-stage anomaly.
    • PSLV C62 / EOS N1: Third-stage anomaly in January 2026.
    • GSLV F15 / NVS 02: Orbit-raising manoeuvres failed.
      • Failure analysis reports for these missions have not been made public.

    What Comes Next?

    • September 2026: GISAT 1A
    • November 2026: NVS 03
    • NVS 03 is expected to restore NavIC to the 4-satellite minimum for standalone positioning.
    • Meanwhile, Indian armed forces continue using NavIC alongside GPS, Galileo and GLONASS.

    “[2018] With reference to the Indian Regional Navigation Satellite System (IRNSS), consider the following statements :
    1. IRNSS has three satellites in geostationary and four satellites in geosynchronous orbits.
    2. IRNSS covers entire India and about 5500 sq. km beyond its borders.
    3. India will have its own satellite navigation system with full global coverage by the middle of 2019.
    Which of the statements given above is/are correct ?
    (a) 1 only
    (b) 1 and 2 only
    (c) 2 and 3 only
    (d) None
    Answer: (a)”

  • Claude AI Gets Global Watermarks to Prove What’s AI-Generated

    Why in the News

    Content generated by Claude will carry a machine readable marking, after Anthropic signed the transparency Code of Practice under Article 50(2) of the European Union Artificial Intelligence Act. The change extends watermarking from images and video to text itself, where the mark travels with copied text and detection is not reliable. The obligation arises from one regional law but the rollout is global.

    What is Anthropic’s new watermarking system?

    1. Trigger: The policy was introduced after Anthropic signed the EU AI Act’s Article 50(2) Code of Practice on Transparency of AI Generated Content.
    2. Two forms of marking: Watermarks are embedded in text content produced by Claude. Signed provenance metadata is attached to supported files in formats such as .svg, .png and .jpg.
    3. Applied at the model level: The text watermark is invisible to users. Anthropic has confirmed that it will not affect Claude’s response.
    4. Persistence: The watermark is part of the text, so it travels with the text when it is copied and pasted elsewhere, and may persist through some editing.
    5. Coverage of surfaces: Output from the Claude Platform (API), Claude, Claude Code, Claude Cowork and Claude Tag is set to carry the embedded watermarks. The same applies when Claude models are accessed through AWS, Google Cloud and Microsoft Foundry.
    6. Detection still incomplete: Anthropic is still working on letting external parties detect the markings, and the rollout announcement did not reveal full technical details.

    What is Article 50(2) of the European Union Artificial Intelligence Act?

    1. Substance: It requires providers of AI systems that generate synthetic text, audio, image or video to mark their outputs in a machine readable format and make them detectable as artificially generated.
    2. Code of Practice route: Signing the associated Code of Practice is the voluntary compliance instrument through which providers demonstrate that they meet the transparency duty.

    What is signed provenance metadata?

    1. About: It is a cryptographically signed record attached to a file that states the file’s origin and the tool that produced it, so a later viewer can verify where it came from.
    2. Weak point: The record is stripped when the file format is converted, which breaks the chain of verification.

    Why does watermarking text change the stakes for ordinary users?

    1. Everyday written work is now in scope: Professional emails, personal messages, school assignments and workplace deliverables that could once pass as human made may carry an AI watermark.
    2. Marginal AI involvement still marks the file: The mark can attach even where Claude’s involvement was close to negligible.
    3. Second hand exposure: A human made file that is proofread, translated, summarised or converted by someone else using Claude can still carry a mark in the final output.
    4. Non users are exposed: A person who never uses the tool can end up holding marked text produced by a collaborator, which has put non users on edge alongside users.
    5. Workflow effect: Millions of customers are reconsidering their use of AI tools and debating at what point human content becomes AI content.

    Why does the mark not settle the question of authorship?

    1. Both error types admitted: Detecting a Claude mark does not confirm that the work was created by AI. The absence of a mark does not confirm that the work was fully human made.
    2. Short text: Short text lengths can throw off the result, since a watermark needs sufficient text to be carried.
    3. Post processing edits: Content changes made after Claude processed the text can degrade the signal.
    4. Format conversion: Metadata is stripped when a file format is converted, removing the provenance record for images and documents.
    5. Unsupported surfaces: Use of a Claude offering that does not yet support AI marking leaves the output unmarked.

    What new risks has the announcement itself created?

    1. A removal market: Multiple dubious websites offering watermark “removal” or “clean up” services came online within days of the announcement.
    2. A repeat of the detector cycle: The earlier rise of AI text detectors was followed by AI text humanisers built to deceive those same detectors.
    3. Reputational damage already recorded: Detector outputs have been involved in cases leading to cancelled book deals and social media trolling for authors and bloggers.
    4. Tool quality: AI text detection tools remain experimental, fallible and prone to errors, yet are treated as evidence.
    5. Credential risk: Users now face the prospect that their own tool damages their professional credentials.

    Why do watermarks work for images but not yet for text?

    1. Images and video are the solved case: Watermarks give regulators, fact checkers and journalists a reliable way to verify the origin of an image or video and trace it to a specific provider.
    2. Text is not: Accurately detecting AI generated text remains uncharted territory, so the same verification logic does not transfer.
    3. Circulation outruns labelling: AI generated content is circulated thousands of times on social media unchecked, as content moderation rules have been loosened across the Meta family of apps and X.
    4. Users do not look: The average internet user scrolling on a phone misses even visible AI watermarks, and an invisible mark is weaker still.
    5. Regulator dependence: A tangible reduction in misinformation and deepfakes requires technology providers and regulators to act together, not a marking standard alone.

    Challenges to AI content watermarking

    1. Adversarial removal: Paraphrasing, translation and dedicated stripping tools defeat statistical text watermarks. e.g. the removal and clean up websites that appeared within days of the Anthropic announcement.
    2. No interoperable standard across providers: A mark from one model tells nothing about content from another, so an unmarked file proves nothing. e.g. the Coalition for Content Provenance and Authenticity (C2PA) standard is adopted by some providers and open source models remain outside it.
    3. False accusation of students and writers: Detector outputs are used as disciplinary evidence despite admitted error rates. e.g. OpenAI withdrew its own AI Text Classifier in July 2023 citing low accuracy.
    4. Open weight models cannot be compelled: A provider level obligation does not reach models that run on a user’s own machine. e.g. freely downloadable open weight models can generate unmarked text offline.
    5. Jurisdictional mismatch: A duty created by one region’s law governs the provider, not the harm suffered elsewhere. e.g. an Indian user injured by unmarked synthetic content depends on a European regulator’s enforcement.
    6. Labelling does not stop the harm: A deepfake remains persuasive even when correctly labelled, because the first viewing shapes belief. e.g. the November 2023 deepfake video of an Indian film actor circulated widely before any advisory was issued.

    Conclusion

    A transparency duty designed for synthetic images and video has been extended to text, where detection is unreliable and the mark attaches to work that may be substantially human. The result is a signal that users cannot see, verify or contest, carrying real reputational consequences. Labelling will reduce misinformation only if detection tools become accurate and platforms act on the marks, neither of which is settled.

    Artificial Intelligence Governance in India

    1. About: AI governance covers the rules on how AI systems are built, trained, deployed and labelled, and who is liable when they cause harm.
    2. No dedicated statute: India regulates AI through existing law and subordinate rules rather than a single AI Act, unlike the European Union’s risk tiered model.
    3. Scale: India has one of the largest AI talent pools and developer bases globally and is among the largest markets for consumer AI applications.
    4. Institutional anchor: The Ministry of Electronics and Information Technology (MeitY) is the nodal ministry, working through the IndiaAI Mission and advisories to intermediaries.
    5. Global positioning: India hosted the AI Impact Summit in New Delhi in February 2026, the successor to the AI Safety Summit series, and is a founding member of the Global Partnership on Artificial Intelligence (GPAI).

    Laws and Rules Governing AI Generated Content in India

    1. Information Technology Act, 2000: The parent statute for electronic records, intermediary liability and cyber offences.
    2. Section 79 grants intermediaries safe harbour subject to due diligence, which is the hook for content labelling duties.
    3. Section 66D penalises cheating by personation using a computer resource, used against deepfake impersonation.
    4. Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021: Impose due diligence, grievance redress and takedown timelines on intermediaries and significant social media intermediaries.
    5. Amendment Rules on synthetically generated information, 2026: Require platforms to label synthetically generated information prominently and to obtain user declarations on whether uploaded content is synthetic.
    6. Digital Personal Data Protection Act, 2023: Governs processing of personal data, including data used to train and prompt AI models, with consent and purpose limitation duties.
    7. Bharatiya Nyaya Sanhita, 2023: Covers forgery, defamation and obscenity offences that synthetic media can constitute.
    8. Copyright Act, 1957: Governs authorship and infringement questions raised by training data and machine generated output.

    Back2Basics: European Union Artificial Intelligence Act

    1. What it is: The world’s first comprehensive horizontal law on artificial intelligence, adopted by the European Union.
    2. Entry into force: 1 August 2024, with obligations applying in phases.
    3. Approach: A risk based classification into unacceptable risk, high risk, limited risk and minimal risk, with duties scaled to the tier.
    4. Prohibited practices: Social scoring by public authorities, untargeted scraping of facial images and manipulative techniques exploiting vulnerabilities.
    5. Article 50: Sets transparency obligations for AI systems that interact with people or generate synthetic content, including machine readable marking of outputs.
    6. Extraterritorial reach: It binds providers placing systems on the EU market irrespective of where they are established, which is why compliance measures are rolled out globally.

    Government Initiatives

    1. IndiaAI Mission: Approved in March 2024 with an outlay of about Rs 10,371.92 crore, built on seven pillars covering compute capacity, innovation centre, datasets platform, application development, future skills, startup financing and safe and trusted AI.
    2. Safe and Trusted AI pillar: Funds work on deepfake detection, algorithmic bias audits and AI governance frameworks, and underpins the proposed AI Safety Institute.
    3. National Strategy for Artificial Intelligence, 2018: NITI Aayog’s framework identifying healthcare, agriculture, education, smart mobility and smart cities as focus sectors under the AI for All approach.
    4. Bhashini: The National Language Translation Mission building open speech and translation datasets across Indian languages.
    5. Responsible AI for Youth: A skilling programme for government school students to build AI literacy at scale.
    6. Digital India Act consultations: Proposed successor to the Information Technology Act, 2000, intended to address emerging technologies including AI and deepfakes.

    Key Facts about AI Content Provenance

    1. C2PA: The Coalition for Content Provenance and Authenticity is the main cross industry technical standard for attaching tamper evident provenance to media files.
    2. SynthID: Google’s watermarking system for AI generated images, audio, video and text.
    3. Deepfake: Synthetic media in which a person’s likeness or voice is replaced or generated, typically using generative adversarial networks or diffusion models.
    4. Turing Test: The 1950 benchmark for machine indistinguishability from a human, now inverted by the problem of detecting machine authorship.
    5. GPAI: The Global Partnership on Artificial Intelligence was launched in June 2020 with India as a founding member, and India held its chair in 2024.

    Challenges in AI Governance in India

    1. No binding statutory framework: India governs AI through advisories and subordinate rules that carry weaker enforceability than a statute. e.g. the March 2024 MeitY advisory on under tested AI models was revised within weeks after industry objections.
    2. Compute dependence: Frontier model training depends on imported accelerators and foreign cloud capacity. e.g. the IndiaAI Mission empanelled over 18,000 graphics processing units in its first round in January 2025 to close this gap.
    3. Data protection enforcement capacity: The Data Protection Board must supervise a very large volume of processors with limited staff. e.g. the Digital Personal Data Protection Act, 2023 rules were notified only in November 2025, years after enactment.
    4. Copyright and training data disputes: Ownership of material used to train models is unresolved in Indian law. e.g. the news agency ANI’s suit against OpenAI in the Delhi High Court filed in November 2024.
    5. Election integrity: Synthetic audio and video can be deployed at scale during compressed campaign periods. e.g. AI generated voice clips of political leaders circulated during the 2024 Lok Sabha campaign.
    6. Algorithmic bias in public service delivery: Models trained on unrepresentative data misclassify beneficiaries. e.g. facial authentication failures for manual workers under Aadhaar based attendance systems.
    7. Skill and audit gap: India lacks a trained cadre of independent AI auditors to test high risk deployments. e.g. no statutory conformity assessment body exists comparable to the notified bodies under the EU AI Act.

    Way Forward

    1. Enact a risk tiered statute: Replace advisory based governance with a law that classifies AI uses by risk and fixes provider and deployer liability.
    2. Mandate interoperable provenance: Require adherence to a common content credential standard so a mark from one provider is readable by all platforms.
    3. Build public detection capacity: Fund an independent testing facility to benchmark deepfake and text detectors and publish accuracy rates.
    4. Protect against false accusation: Bar educational institutions and employers from acting on detector output alone, and require corroborating evidence.
    5. Expand sovereign compute: Scale domestic graphics processing unit capacity and public datasets so Indian models are not fully dependent on foreign infrastructure.
    6. Strengthen platform duties: Require prominent labelling at the point of display, not only in file metadata, and fix takedown timelines for unlabelled synthetic media.
    7. Invest in digital literacy: Run sustained public campaigns so users check provenance labels rather than react to content at first sight.

    “[2023, GS3, 10 marks] Introduce the concept of Artificial Intelligence (AI). How does AI help clinical diagnosis? Do you perceive any threat to privacy of the individual in the use of AI in healthcare?”